ANTIQUITIES
SOUTHERN"
GEORGIA TRIBES.
BY
CHARLES C. JOKES, JK.
NEW YORK: D . A P P L E T O N A N 1) COMPANY
549 & 551 BROADWAY. 1873.
CITAHI/ES C. JONES, JR., Ilio office of the Librarian of Congress, at Washm^to:
THE STATE OF GEORGIA,
THIS VOLUME IS Al'FKCTIONATELY IffSCRI OXE OF HER SONS
PKEFA.CE.
AI/THOTJGII the title intimates that our investiga tions have been directed principally to an examination of> the antiquities of a single State, the present work will be found to embrace within its scope a much more extended field of observation. In prosecuting the pro posed inquiries, it appeared both unnecessary and im proper narrowly to observe the boundary-lines which separate modern States. It will be remembered, more over, that the original grant from the British crown conveyed to the Trustees of the Colony of Georgia a territory greater by far than that now embraced with in the geographical limits accorded to her as a State. A striking similarity exists among the customs, uten sils, implements, and ornaments of all the Southern Indians : consequently, in elucidating the archeology of a region often occupied in turn by various tribes, it seemed appropriate to mention and contrast the
antiquities of Virginia, the Carolinas, Florida, Ala bama, Mississippi, Louisiana, and Tennessee.
Our object lias been, from the earliest and most authentic sources of information at command, to con vey a correct impression of the location, character istics, form of government, social relations, manufac tures, domestic economy, diversions, and. customs of the Southern Indians, at the time of primal contact between them and the Europeans. This introducto ry part of the work is followed by an examination of tumuli, earthworks, and. various relics obtained from burial-mounds, gathered amid refuse-piles, found in an cient graves, and picked up in cultivated fields and on the sites of old villages and fishing-resorts. "When ever these could be interpreted in the light of early recorcled observations, or were capable of explanation by customs not obsolete at the dawn of the historic period, the authorities relied upon have been carefully noted.
The accompanying plans of mounds were prepared from personal surveys, and nearly every typical object used in illustration may be seen in the author's collec tion. Most of these relics were obtained by me in situ. They are now figured for the first time.
To the friends who have kindly aided me in gather ing together a cabinet which so fully and beautifully represents the arts and the manufactures of these
primitive peoples, I here renew my cordial and gratefnl acknowledgments.
Prepared at irregular intervals and in odd mo ments as they could be borrowed from the exacting and ever-recurring engagements of an active profes sional life, these pages, -with their manifest short comings, are offered in the hope that they will, at least in some degree, minister to the information and pleasure of those who are not incurious with regard to the subject of American archseology.
KEW YOEK, April 10, 1873.
C. JOKES, JR.
CHATTER I.
Location of Tribes.--Physical Characteristics of the Southern Indians.--System
of Government.--The Mico,--The Bead War-Chief.--Public Buildings in a
1 Creek Village.--Mode of Warfare.--Office of High-Priest.--Sun-Worsbip.--
Offering of the Stag.--Idol-Worship.-^-Religious Ideas.--The SUTI among the
AT ateljez,--The Cacica of Cutifacliiqiu.--Mausoleum of Titlomeco.--Tombs of
the Virginia Kings,
......
PAGE 1
CHAPTEK II.
OSes of the Conjurer or Medicine-man.--Treatment of the Sick.--Medicinal I'liints.--Towns and Private Houses.--Tenure of Property.--Agricultural Pursuits.--Town Plantations and Private Gardens.--Public Granaries.--Ani mal and Vegetable Food.--Mechanical Labors.--Early Mining inDuke's-Crec-k Valley.--Manufacture of Canoes, Pottery, Copper Implements, Gold, Slyer, Shell, and Stone Ornaments.--Various Implements and Articles of Stone, Rone, and Wood.--Trade Relations, ....'. 58
CHAPTER III.
Hamate and Divorce.--Punishment of Adultery.--Costume and Ornaments.--
Skin-painting and Tattooing.--Manufacture of Carpets, Fcather-sbawla, and
Moccasins.--Weaving, .
.
.
.
.
.
.65
CHAPTER IV.
Music and Musical Instruments.--Dancing.--Game?. --Gambling.--Festivals.--
Divisions of the Tear.--Counting.--Funeral Customs, .
.
90
COjSTTKNTS.
CHAPTER V.
General Observations on Mound-Building.--Bartram's Account of the Georgia
Tumuli.--Absence of Mcgalithie Monuments and Animal-shaped Mounds.--
Distribution of the Ancient Population.--Few Sepulchral Mounds erected
since the Advent of Europeans.--Antiquity of the Tumuli,
. PAGE 118
CHAPTER VI.
Mounds on the Etowah River.--Temple for Sun-worship.--Stone Images.--Fish-
Preserves.--Tumuli in he Valley of Little Shoulder-bone Creek.--Circular
Earthwork on the Head-waters of the Ogeechee.--Stone Tumulus near Sparta,
--Mounds on the Savannah River.--Meeting between the Cacica of the Savan
nah and De So to,
,
.
,
-
,
.
.130
CHAPTER VII.
CHAPTER VIII.
Chunky-Yards.--Elevated Spaces.--Mounds of Observation and Retreat. --Tumuli
on Woolfolk's Plantation.--Sepulchral Tumuli.--Chieftain-Mounds.--Custom
of burying Personal Property with the Dead.--Savannah owes a Monument
to Tomo-chi-chl.--Family or Tribal Mounds.--Cremation,
.
.118
CHAPTER IX.
CHAPTER X.
Stone Graves in Nacooehee Valley and elsewhere.--Copper Implements and the
Use of that Metal among the Southern Indians.--Cane-Matting.--Shell
Drinking-Cups,--Shell Pins.--Age of Stone Graves,--Evidence of Commerce
among the Aborigines, .
.
.
.
.
.
.213
CHAPTER XI.
CONTENTS.
SI
CHAPTER XII.
Grooved Axes.--Hand and Wedge-shaped Axes or Celts.--Perforated and Orna
mental or Ceremonial Axes.--Chisels.--Gouges.--Scrapers.--Flint Knives.--
Avis, or Borers.--Leaf-shaped Implements.--Smooth in g-Stones.--Drift-Im
plements,
....... PAGK 269
GRAFTER XIII.
Agriculture and Agricultural Implements.--Ceremony of the Busk.--Cultivation
of Maize.--Mortars and Pestles.--Crushing-Stones.--Nut-Stones.--Use of
Walnut and Hickory-nut OH, .
.
.
.
.
. 296
CHAPTER XIV.
. Fishing.--Wears.--Nets.--Net-sinkers.--Plummets, .
.
.
321
CHATTER XV. . Diseoidal Stones.--Chungke-Game, ...... 841
\
CHAPTER XVI.
Stone Tubes, .
.
.
.
.
.
.
.
850
CHAPTER XVII.
Stones for rounding Arrow-shafts.--Whetstones or Sharpeners.--Pierced Tablets. --Pendants.--Slung-slones.--Amulets.--Stone Plats.--Mica Mirrors.--Sculp tured Bocks, ........ 360
CHAPTER XVIII. Pipes.--The Use of Tobacco.--Idol Pipes.--Calumets.--Common Pipes.
. 883
CHAPTER XIX. Idol-Worship among the Southern Indians.--Stone and Tcrra-Cotta Images, 413
CHAPTER XX.
Pottery,
.
.
.
-
.
.
.
.
. 441
CHAPTER XXI.
The Use of Pearls as Ornaments among the Southern Indian?,
.
. 46 Y
CHAPTER XXII.
Primitive Uses of Sheila.--Shell-Honey.--Shell Ornaments.--Personal Decorations. --Conclusion, ........ 495
F
LIST OF ILLUSTRATIONS.
PLATE I. (To face page WQ.} , Tumuli and Fish-Preserves in the Etowah Valley, Georgia,
\
PLATE II. (To face page 144.)
Figs. 1 and 2. Tumuli in the A7alley of Little Shoulder-bone Crock.
3. Enclosed Work. * 4. Circular Earthwork on the Head-waters of the Ogeechee.
5. Stone Tumulus near Sparta, Georgia.
PJ.A.TE III. (To face page 152.) Tumuli on the Savannah Eiver, "below Augusta.
PLATS IV-. (To face page 158.) Tumuli on the Ocmulgee River, opposite the City of Macon.
PLATE IV., A. (To face page 160.)
Fig. 1. Skull of a Creek Indian. 2 and 3. Two Views of the Skml of an Ancient Mound-builder.
PLATE V. (To face page 1G8.) Mound on Messicr's Plantation, in Early County.
PLATE VI. (To face page 224.)
JRclics found in Stone Graves in N"acoochec Valley. Fig. 1. Cane Matting.
2-7. Copper Implements. 8 and 9. Shell Pins. 10. Soapstone Pin. 11 and 12. Stone Beads.
LIST OF ILLUSTRATIONS.
PLATE VII. {To face page 252.)
Figs. 1 arid 2. Large Flint Spear-heads. 8 and 5. Flint Daggers. 4. Serrated Flint Sword.
PLATE VIII. (To face page 254.) Figs. 1-12. Typical Forms of Flint Spear-heads.
PILATE IX. (To face page 25G.) Figs. 1-41. Typical Forms of Arrow-points.
PLATE X. (To face page 274.) Figs. 1-7. Typical Forms of Grooved Stone Axes.
8, Stone Adze.
PLATE XI. (To fate page 278.) Figs. 1-6. Typical Forms of Polished Sfc^ne Celts.
PLATE XII. (7b face page 280.) Stone Axe from Tennessee.
PLATE XIII. (To face page 282.)
Figs. 1-5. Typical Forms of Perforated and Ornamental or Ceremonial
Hatchets. - '
6. Hammer-Stone.
PLATE XIV. (TofaeepayeZQQ.)
Fig^ i_4. Stone Chisels. 5-7. Stone Gouges. 8. Bone Gouge. 9-14. Typical Forms of Stone Scrapers.
PLATE XV. {To faze page 290.) Figs. 1--9. Flint Knives and Leaf-shaped Implements.
PJ.ATB XVI.
Fig. 1. Bone Awl. 2--5. Stone Borers. 6-9. Smoothing-Stoiies.
10. Drift Implement.
(To face page 292.)
PLATE XVII. (To fa-ce pags 302.)
Fig. 1. Stone Hoe. 2. Stone Spade. 3-5. Flint Agricultural Implements.
I.I3T OF ILLUSTRATION.
XV
PLATE- XVIII. (To face page 812.)
Figs. 1 3. Stone Mortars. 4-5. Stone Pestles. G and 8. Maize-crush firs or Triturating Stones. 7. Stone upon which Nuts were cracked.
PLATK XIX. {To face page 338.)
Figs. 1-fi. Perforated Stone Net-sinkers.
7-11. Grooved "
"
12. Fishing Plummet.
PLATE XX. (To faze page*S4.) Figs. 1-13. jJiscoidal Stones.
PLATE XXL (To faaepage 858.) Figs. 1-fj. Stone Tubes.
PLATE XXIT. (To f.ice page 36G.)
.Fig. 1. Stone for rounding Arrow-shafts. 2 and 3. Pierced Tablets.
1 4. Slung-stone. 5. Amulet. 0. Stone Plato. 7. Whetstone.
PLATE XXIII. (To face pays 404.) Figs. 1-0. Typical Forms of Calumets.
PLATE XXIV. (To face page 410.)Figs. 1-7. Typical Forms of Common Clay and Stone Pipes.
PLATE XXV. To face paye 430.) Figs. 1-0. Clay Images.
PLATE XXVI. (To face page 432.)
Figs. 1-3. Front, Side and Bear Views of the Stone Image found in the Etowah Valley, Georgia.
PLATE XXVII.
Fig. 1. Bnrial-Crn. 2. Large Earthen Pot. 3 and 4. Jars. 5-7, Pots with Ears. S. Tot with Legs. 9 and 10. BowLs.
(To face page 454.)
LIST OF ILI/tTSTKATIOXS.
PLATK XXVIII. (To face page 4SG.) Fig. 1. Jar.
2. Burial-Urn. 3 and 4. Vessels with Narrow Necks. 5-10. Pottery from Stone Graves of Tennessee.
PiATJ; XXIX. (To face page 433.) !. 1-32. Sherds, showing the Ornuiuentution of Primitive Potte:
PLATE XXX. (To face pay,, 503.) >. 1 and 2. ^Vampum or Shell Money.
3 and 4. Shell Gorgets. 5-7. Shell Pins. 8. The Oliva as a Shell Bend. 9. The Marginella as a Shell Bend. 10-12. Impcrforate Columns of Sea-Shells as Articles of Col 13. Bone Bead. 14-19. Typical Forms of Shell Beads.
WOOD C 17 T S .
ANTIQUITIES
SOUTHERN
CHAPTER I.
Location of Tribe*.--Physical Characteristics of the Southern Indians.--Svstcm of Government.--The Mico.--The Head AVar-Chief.--Public Buildings in a Greek Village.--Mode of Warfare.--Office of High-Priest.--Sun-Worship.--
I Offering of the Stag.--Idol-Worship.--Religious Ideas.--The Sun among (lie Natchcz.--The Caeiea of Cutifaehiqui.--Mausoleum of Talomcco.--Tombs of the Virginia Kings.
BY letters patent, dated tlie 9th of June, 1732, King George II. incorporated the trustees for estab lishing the colony of Georgia in America, and con veyed to them and their successors " seven-eighths of all that territory lying between the Savannah and Alatamaha Rivers, and westwardly from the heads of the said livers respectively, in direct lines, to the sonth seas." In this alienation were embraced all isl ands within twenty leagues of the const. Including a large portion of the present States of Alabama and Mississippi, this grant claimed an extension, in a west erly direction, as indefinite as "was then the geographi cal knowledge of the region intended to be compi'ised in the royal feofmerit.
Of the Indian nations, east of the Mississippi River, occupying and living adjacent to this territory about
L
12
ANTIQUITIES OE THE SOUTIIEEB" INDIANS,
the beginning of the eighteenth century, the dominant peoples were tlie lichees, Lower, Middle, and Upper Creeks--constituting the formidable MUSCOGEE CON FEDERACY--the Yamasees, the Cherokees, the Chickasaws, the Choctaws, the I^atcliez and the Seminoles. ' East of the Savannah Kiver resided the Catawbas, the Savannahs, and the Westoes--the latter tribe includ ing the Stonoes and the Edistoes--cruel and hostile peoples, between whom and the Carolina colonists early and prolonged warfare ensued. The Yamasees are mentioned by Governor Arehdale a as living about eighty miles from. Charleston, and extending their hunting excursions nearly to St. Augustine. This was in 1G95. Between, the "Westoes and the Savan nahs--both potent tribes and numbered bv " many thousands" -- a violent civil strife arose, in conse quence of which they were greatly reduced in popula tion and resources. This contest resulted in the final overthrow and expulsion of the ^Vestoes--" the more cruel of the two "--the Savannahs continuing " good friends and useful neighbors to the English." ' Small pox and other unusual sicknesses are said, at an early period of the English colonization of Carolina, to have wrought sad havoc among the natives.
Surveyor-General Lawsoii describes the Savannahs as a " famous, warlike, friendly nation of Indians ]iv-
l History of East ana W>sst Florida," pp. 5!)-C7, 71.
THK YAMASEES, Tj CHEFS AKD CREEKS.
3
ing to the south end of Ashly River." * They prob ably derived their name from the river whose banks they inhabited, and it is Mr. Gallatiii's opinion 2 that they and the, Yamasees were one and the same people, the latter being the true Indian name.
These Yamasees and their confederates were, ill 1715, routed by Governor Craven and driven across the Savannah River into the arms of the Spaniards in Florida. -/It is not improbable that the.Yam a ci-mvs. who were occupying the present site of the city of Savarinah whei General Oglethorpe landed and estab lished the colony of Georgia, were a remnant of this tribe. Among the allies of the Yamasees the lichees vrere numbered, and they, too, after this signal discom fiture, contented themselves with a residence in the everglades of Florida. Theirs, of all the Indian lan guages of this region, was the most uncouth and gut tural. Bartram asserts that their national language was radically different from the Miiscogulgee tongue. He was informed by the traders that their dialect was the same as that of the Shawiiees. Although at one time confederated with the Creeks, they refused to mix with them and excited the jealousy of that whole nation.
The Chickasaws at one period occupied the left bank of the Savannah River opposite Augusta.*
About the date of the colonization of Georgia, the territory of the Creek Confederacy--including lands inhabited, by the Semiuoles--was bounded on the west by Mobile River and by the ridge separating the waters of the Tombigbee from those of the Ala bama (the latter being the contested boundary-line be-
1 " History of Carolina," p. 42. London, 1714. 3 Synopsis of the Indian Tribe*, " Arehieologia Americana," vol. ii., p. 84. a Iluywood'ri " Aboriginal History of Tennessee," p. 290.
4
ANTIQUITIES OP THE SOITTHERJT ISDIAisS.
tweeii the Creeks and tlie Choctaws), on the north by the Cherokees, 011 the northeast by the Savannah River, and on every other side by the Atlantic Ocean and the G-nlf of Mexico. It is believed, at the end of the seventeenth century, that, sonth of the thirty-fonrth degree of north latitude the Greets occupied the eastem as well as the western bank of the Savannah. It cannot now be ascertained with certainty when the consolidation of this confederacy was effected. "It is probable," says Mr. Gallatin, " that the appellation of Appalachians was geographical, and applied to the Indians living on the Appalnchicola or Chattahoochee River, as the name of Creeks seems to have been given from an early time to those inhabiting1 generally the country adjacent to the Savannah River." Of the Creek Confederacy, by far the most numerous and powerful nation was the IVEnscogee.1 The Hitehittees, who resided on the Chattahoochee and flint .Rivers, although a distinct tribe, spoke a dialect of the JMiiscogee. The Seiniiioles, or Isty-seniole (wild men), inhabiting the peninsula of Fiorida, -were pure jSluscogees, and received that name because they subsisted principally by hunting and devoted but little attention to agriculture.*
"When questioned as to their origin, the Muscogees responded that the prevailing tradition among them was, that their progenitors had issued out of a cave near the Alabama, River, The account given by tlie Hitchittees of their beginning wa,s scarcely less fanci ful. They claimed that their ancestors had fallen from the sky.
THE CIIACTAWS, UCIIEES
"The Chactaws." says Captain Romans, "have told me of a hole between their nation and tlie Chicasaws, out of which their whole, very numerous nation, -walked, forth at once, without HO much as warning any neigh bor." '
The Uchees and the x^atehez both acknowledged allegiance to the Creek Confederacy. The original seats of the Uchees are thought to have been east of the Coosa, and probably of the Chattahooc-hee. They declared themselves the most ancient inhabitants of the cou itry, and it has been sugg'ested that they were the peoples called Appalach.es by the historians of De Soto's expedition. Their country wras mentioned as a land abounding in towns and subsistence. Marly in the eighteenth century, they occupied the western bank of the Savannah. River; and, as late as 1736, claimed the country both above and below Augusta. The name of at least one creek in Columbia County perpetuates at once their memory and. the fact of their former occupancy of this region.
A residue of the iSTatchez forsook their old habitat on the banks of the Mississippi, and, journeying east ward, associated themselves with the Creeks less than one hundred and tifty years ago. The principal towns of the Creeks were Cussetah, Cowetah, Tukawbatchie, and. Oscoochee." The 'Mnscogee, the Hitchittee, the Uchee, the jSTatehez, and the Alibamoii or Coosada, were the principal languages spoken by the various tribes composing the Creek Confederacy. On the 12th of March, 1733, General Oglethorpe mentions the Lower and Upper Creeks, and the Uchees, as the three
'"Coneiao Natural History of Bust and West Florida," p. 68. lilt.
2 "Arcbffiologia Americana," vol. ii., p. 95.
Xew York,
(>
ANTIQUITIES OF TILE SOUTHERN INDIAXS.
most powerful Indian nations in Georgia between tlie mountains and tlie coast. The Lower Creeks consisted of nine towns or cantons, and tlieir warriors were esti mated by him at one thousand. The military strength of the Upper Creeks he computes at eleven hundred moil capable of bearing arms, while it was supposed that tlie Uchees were at that time unable to bring into the field more than two hundred bow-men. This esti mate is evidently too small, and was vaguely formed. T)e Braluu, 1 at a later date, reckons the population of the Upper and Lower Creeks at fifteen thousand men, 'women, and-children, and rates their warriors and gun men above three thousand. To Colonel Hawking we are indebted for a very valuable sketch of the Creek country in 1*798 and 1799.
The Creeks are described as powerful warriors, great politicians, and full of jealousy. They were a terror to the Cherokees and to the various Indian na tions with whom they waged ceaseless wars.'
Captain Romans 4 enumerates remnants of the Cawittas, Talepoosas, Coosas, Apalachias, Conshacs, or Coosades, Oakmulgis, Oconis, Okchoys, Alibamons, Watchez, Weetumkus, Pakanas, Taensas, Chaesihoomas, Abekas, and of other tribes, whoso names lie did not recollect, all calling themselves Jlfnscokees, and consti tuting what was known as the Creek Confederacy.
"The territories of the Cherokees, Chelakees, or more properly, Tsalakies," says Mr. GaUatin," extended
1 "History of the Province of Geoi-gia," etc., p. 55. Wormsloe, 1S19. 2 " Collections of tlie Georgia Historical Society," vol. iii., part 1., p. IS, et x'-ff. Savannah, 184S. ', also, "A Voyage to Georgia," begun in 'Jie year IVa.j, by praiicis Jloore, p. 61. London, 1741. :1 " History of tlie Province of Georgia," by Do Brahtn, p. 55. Wormsloe, 18-19. Adair's " History of the North-American Indians," p. 257, ci setj. Tendon, 1775. " Travels," etc., by William Eartram, p. 461, *'. seg. London, 1722. 4 " Concise Xataral History of East and West Florida," etc., p. 9O. New York, 1775.
TERRITORY Of THE CHEROKHES.
t
nortli and south of the southwesterly continuation of the Appalachian Mountains, embracing oil the north the country on Tennessee or Cherokee Kiver and its tributary streams, from their sources down to the vi cinity of the Muscle Shoals, where they were bound ed on the west by the Ohicagas. The Cumberland mountain may bo considered as having been the bound ary on the north; but, since the country has been known to us, no other Indian nation, except some small bands of Shawiioes, had any settlement between that mount, in and the Ohio." On the west side of the Savannah, the Cherokees were confronted on the south by the Creeks, the division-line being Broad River and generally along the thirty-fourth parallel of north lati tude. East of the Savannah, their original seats em braced the upper waters of that river, of the Sautee, and, probably, of the Yadkin, but could, not have ex tended as far south as the thirty-fourth degree of north latitude. They were bounded on the south, probably, by Muskhogee tribes in the vicinity of the Savannah, and, farther east, by the Catawbas."
Between the Shawnoes and the Cherokees prolonged strife occurred, which resulted in the expulsion of the former from the country south of the Ohio. T\rith the Creeks also the Cherokees were constantly at variance. When in 1730 the whites interposed their good offices to bring about a pacification between the Tnscaroras and the Cherokees, the latter responded: " Wo cannot live without war; should we make peace with the Tuscaroras, with whom we are at war, we must imme diately look out for some other with whom we can be engaged in. our beloved occupation." 2
s Haywood's "Natural and Aboriginal History af Tennessee^" p. 238. Xasbville, 1828.
8
ANTIQUITIES OF THE SOUTHERN" INDIANS.
The history of the Ckerokees is marked 1:>J contintied and prolonged struggles. Their country being strong, their men trained to anas, and the integrity of the nation at all times wonderfully preserved, these peoples do not appear, in their territorial possessions, to have been materially injured by their frequent contests with adjacent tribes. In 1762 Adair estimated the number of their warriors at three thousand two hun dred, and adds, he was informed that, forty years before, they had at least six thousand men capable of bearing arms.'
In perpetuating his impressions of the PHYSICAL ciiAUACTHiiiSTtcs of the Southern Indians, Mr. I->artram * writes : " The males of the Cherokees, Muscogiilgees, Siminoles, Chicasaws, Chactaws, and confederate tribes of the Greeks, are tall, erect, and moderately robust; tlieir limbs well shaped, so as generally to form a per fect human figure; their features regular and counte nance open, dignified, and placid ; yet the forehead and brow so formed as to strike you instantly with heroism and bravery ; the eye, though rather small, active and full of fire ; the iris always black, and the nose com monly inclining- to the aquiline. Their countenance and actions exhibit an air of magnanimity, superiority, and independence. Their complexion of a reddish brown or coppe.r color ; tlieir hair long, lank, coarse, and black as a raven, and reflecting the like lustre at different exposures to the light. The women of the Cherokees are tall, slender, erect, and of a delicate frame ; their features formed with perfect symmetry, their counte nance cheerful and friendly ; and they move with a becoming grace and dignity.
1 " History of the American Indians," p. 227. London, 1Y75. -" Travels through ^orth and South Carolina, Georgia," etc., p. 481. Lon don, 1792.
THE MUSCOGUXGHES ASD CITEROKEES.
9
" The Muscogulgee women, though remarkably short of stature, are well formed ; tlieir visage round, features regular and beautiful, the brow high and arched ; the eye large, blade, and languishing, expressive of modesty, diffidence, and bashfillness; these charms are tlieir de fensive and offensive weapons; and they know very well liow to play them off; and, nnder cover of these alluring graces, are concealed the most subtle artifices; they are, however, loving and affectionate ; they are, I believe, the smallest race of women yet known, seldom above five feet high, and I believe the greater number never arrive to that stature ; tlieir hands and feet not larger than those of Europeans of nine or ten years of ago; yet the men are of gigantic stature, a full size larger than Europeans ; many of them above six feet, and few under that, or live feet eig'ht or ten inches. Tlieir complexion much darker than any of the tribes to tlie north of them that I have seen. This descrip tion will, I believe, comprehend the Muscogulges, their confederates, the Clmctaws, and, I believe, the Chicasaws (though I have never seen their women), except ing some bands of the Siminoles, Uches, and Savaunucas, who are rather taller and slenderer and their complexion brighter.
" The Cherokees are yet taller and more robust than the Muscogulges, and by far the largest race of men I have seen; their complexions brighter and somewhat of the olive cast, especially the adults ; and some of their young women are nearly as fair and blooming as Euro pean women.
"The Cherokees, in their dispositions and manners, are grave and steady; dignified and circumspect in their deportment; rather slow and reserved in conver sation ; yet frank, cheerful, and hninane; tenacious of
10
ANTIQUITIES OP THE SOUTHERN DTDIAH3.
the liberties and natural rights of man; secret, delib erate, and determined in their councils; honest, just, and liberal, and ready always to sacrifice every pleasure and gratification, even their blood, and life itself to de fend their territory and maintain their rights. . . .
" The national character of the IMuscogulges, when considered in a political view, exhibits a protraiture of a great or illustrious hero. A proud, haughty, and arrogant race of men, they are brave and valiant in war, ambitious of conquest, restless and perpetually exercising their arms, yet magnanimous and merciful to a vanquished enemy when he submits and seeks their friendship and protection; always uniting the van quished tribes in confederacy with them: when they immediately enjoy, unexceptionably, every right of free citizens, and are, from that moment, united in one common band of brotherhood. They were never known to exterminate a tribe, except the Yamasees, who would never submit on any terms, but fought it out to the last, only about forty or fifty of them escaping at the last decisive battle, who threw themselves under the protection of the Spaniards, at St. Augustine. . . . The Muscogulges are more volatile, sprightly, and talk ative, than their northern neighbors, the Oherokees."
The SYSTEM op GOVEEIOIEKT obtaining among these Southern nations seems, in its general features, to have been quite similar. In the Muscogulgee confederacy every town or village was regaz'ded as an independent nation or tribe having- its own mico or chief. In the soil and in the hunting privileges of the region each inhabitant had an equal right. Private property in habitations and in planting-grounds, however, wns conceded and respected.
THE MICO AND CHEAT WAR-CHI"EF.
11
The MICO * "\vas considered the first man, In dignity and power, in his nation or town. lie was the su preme civil magistrate, and presided over the national council. His executive power was not independent, however, of the council, which convened every day, in the forenoon, in the public square. This office of mico or king was elective. The advancement to this su preme dignity was always conferred upon the person most worthy of it.
Next in the order of dignity and power was the GKKAT W^_E-CHIEF. He led the army. In council his seat was nearest the mico, on his left, and at the head'of the most celebrated warriors. On the right of the mico sat the second head-man of the tribe, and '..below him the younger -warriors of the nation.
"When, assembled in. the Great Rotunda, or "Winter Council-House, for the purpose of deliberating upon matters of general concern, the most profound respect and homage were paid by every one to the mico. To him the members of the council bowed very low, almost to his feet, when the cup-bearer handed, him the shell filled with the black-drink.2 This decoction of the leaves and tender twigs of the cassine or ilex y^t/pon was freely used "by the natives upon occasions of sol emn deliberation. Being- a most active and powerful diuretic, its purgative influences were invoked to free their bodies from all hinderaiice to thought; and, thus prepared for careful discussion, they entered upon the consideration of the important matters presented for the action, of council. De Bry presents us with a spir ited sketch of the kins' and warriors in convention
12
ANTIQUITIES OF THE SOUTHERN INXHAKS.
assembled, drinking freely of tliis oassine from shellcups and listening to the. animated address of one of the principal men. ^Vlien out of the council-house, and unemployed in public affairs, the intercourse be tween, the mico and the common, people was cordial and free from restraint. If we may credit the repre sentations of De Bry,1 no little ceremony was observed when, the king-s and queens of the Florida tribes ap peared in public. The mico alone had the disposal of the corn and fruits collected in the public granary. These general storehouses, circular in form.--their walls constructed of stone and. earth, and their roof's fashioned with tlie brandies of trees, grass, clay, and palmetto-leaves--were located in the -neighborhood of streams and in retired spots where they were protected from the direct rays of the sun. They were built and furnished by the common labor of the tribe, and in them were stored corn, various fruits, and the flesh of fishes, deer, alligators, snakes, dogs, and other animals, previously smoked and dried on a scaffold.2
"With the first fruits of the season was the king complimented. It was his province to give audience to ambassadors, deputies, and strangers, and to him were pxiblic presents offered. He alone had the privilege of giving a general feast to ail entire village., on which occasion the king's standard was displayed in front of his house, a nag hoisted in the public square, drums boat about the town, and the inhabitants busily en gaged in painting and dressing themselves for the festivities. In the sixteenth century, the Florida -war riors, -when about to set out on a ho.stile expedition,
THE OFFTC15 Oil1 KIKG.
13
assembled round their kino-, who, taking a dipper of water and sprinkling them, exclaimed, " As I have scat tered this water, so do you cause the blood of your enemies to flow freely." Then, with water from another vessel, extinguishing a iire kindled in the cir cle, he added, " As I have put out this flame, so may you vanquish, and destroy your antagonists." '
It would appear that, on some occasions, the king, wdieii about to enter into battle, was borne upon a platform elevated upon the heads and shoulders of his men."
The care and protection of widows, whose hus bands had fallen ill battle or perished by disease, devolved upon the king."
Capital punishment was meted out ill the pres ence of the mico and council seated in a semicircle, the victim kneeling in the centre, and the executioner, his left foot upon the back of the criminal, with a stout, paddle-shaped club made of hard wood, striking him upon the top of the head with such violence as to split the skull."
The custom obtained among some of the Southern nations of sacrificing to the king the first-born male child."
The office of king was for life, or during good be havior.
It cannot be denied that to the kingly office, among most of the Southern tribes, appertained, des potic powers. Especially was this the fact at the period of our first acquaintance with the form of gov ernment dominant among these peoples. By at least
1 ''Brevis Nim-atio," plate xi. - Ibid., plate xiii. 3 Ibid., plate xsiii. 4 Ibiit., plate xxxii. * Ibid., plate xxxrii.
14
AWriQTJITIKS OF THE SOUTHERN INDIANS.
one of the early historians are we assured that, ill sa luting a cacique, the subject need gestures, somewhat modified in degree, but similar in form to those em ployed in the adoration of the sun. The intimation is that in hia person and position were recognized a su periority, a dignity, and an authority near akin, but subordinate to those which inhered in the celestial luminary--the most potent and admirable representa tive of the goodness and supremacy of the Great Spirit.
Colonel Hawkins * thus epitomizes the duties of the Creek mico in 1798: The mico of the town super intends all public and domestic concerns, receives all public characters, hears their talks, lays them be fore the town and delivers the talks of his town. The mico of a town is always chosen from some ojv* J* ily. The mico of Tuck-au-bat-che is of the eagle tribe (Lum-ul-gee). After he is chosen and put on his scat, he remains for life. On his death, if his nephews are At for the office, one of them takes his place as his successor; if they arc unfit, one is chosen of the next of kin, the descent being always iu the femaleline.
When a mico, from age, innrmity, or any other cause, wants an assistant, he selects a man who ap pears to him the best qualified, and proposes him to the councillors and great men of the town, and, if he is approved by them, they appoint him as an assistant in public aAairs.
The mico, councillors, and warriors, meet every day in the public square, sit and drink a-cee--a strong decoction of the cassine yupon, called by the traders
* " Skutch of the Greek Country." Collections of tho Georgia Historical Soci ety, ?oL 111., part 1., p. 60. Suvaonmh, 1848.
WAK, PEACE, PUBLIC BUILDINGS.
15
lilaok'drinlc--talk of news, the public and domestic concerns, smoke their pipes and play thla-chal-litch-cau. They have a regular ceremony for making' as well as delivering the a-cee to all who attend the square. In all transactions which require secrecy the rulers meet in the chooc-ofau-thluc-co--the -rotunda or (fssemblyroom called by the traders Jiot-Jiouse--kindle the spiral fire, deliberate, and decide. When they have decided oil any case of death or whipping, the mico appoints the warriors who are to carry it into effect, or he gives the judgment to the great warrior (tustuimuggeethluc-co) and leaves to him the time and mariner of executing it.
^Var is always determined on by the great warrior. If the mico and warriors are of opinion that the town has been injured, it is the province of the former to lift the hatchet against the offending nation. Even after a declaration of Avar, however, the mico and councillors may interpose and proceed to adjust the misunderstanding by negotiations.
Peace is concluded by the mico and councillors, and peace-talks are always addressed to the cabin of the rnico. It is the privilege of the mico and coun cillors to fix the precise time for the celebration of the annual festival of the Boos-ke-tan.
Of the buildings which formed the public square in the Creek villages, the first in rank: was the mic-ulgee-in-too-pau, or mico's cabin. It fronts the east. The centre of this building is occupied by the mico, the rig'ht division by the inic-u^-gee and the council lors, and the left division by the people second in com mand, who have the direction of the public works appertaining to the town.
Second in rank is the tus-tun-iiug-ul-gee-in-too-pau,
1(5
ANTIQUITIES OI' THE SOUTIIEKS INDIAKS.
or warrior's cabin. Tliis fronts the south. At the west end of this cabin sits the head-warrior. In this division are seated also the great warriors. The next in rank sit in the centre division, and the young war riors in the third. These warriors rise by merit, and the great-warrior is selected by the mice and coun cillors as the most noted of all the warriors. .The cabin of the beloved men--is-te-chaguc-ul-gee.-in-toopau--fronts north and is erected for the accommoda tion of those who have been war-lenders and who have rendered themselves distinguished by a long course of valuable public serx'iee. I^ast in rartk is 'hut-te-mau-lmg-gee-in-too-pati--the cabin of the young people and their associates. This fronts the west. To these may be added the chooc-ofao-thluc-co--the ro tunda, or assembly-room, called by the traders Jiot7iouse. In the centre of this is the spiral fire. This is the assembly-room for all people, old and young. Here they congregate every night, and amuse them selves with dancing, singing, or conversation. In this building sometimes, in very cold weather, the old and linked sleep.
In the absence of the mico, the OpvEAi7 "\VAi;-Ciri.KF ' represented him in council, and his voice was of the greatest weight in military affairs. His authority was independent of the mico, although, should the mico enter upon a military expedition, he was entitled to the command. Subordinate to the great war-captain were leaders of parties--elderly men distinguished for valor, strategy and intrepidity. Of such were their dignified and venerable councils composed.
Having by fasts and purifications prepared them selves for the expedition, having consulted the higli-
1 "Bartram's Travels," p. 494. Lomion, 1792.
MODE OF WARFARE.
17
priest,1 with, regard to the success of the enterprise, and obtained from him a favorable response, fantastically painted and plumed, each carrying a small bag of parched corn, and armed -with a long boMr and. quiver of arrows suspended from the right hip, and frequently with a formidable club made of hard wood, and a spear,* the warriors set off from the village with a great noise and defiant shouts. The head-warrior, talcing the lead, was followed by the rest in single file. "When. near the hostile town or in the vicinity of the spot where a meeting with the enemy was anticipated, the most profound silence and careful circumspection were observed. Their conduct then resembled the action of . the concealed lynx waiting for an opportunity to ',-*-*-*-" nee upon its prey in an unguarded moment. A. su^.'en attack, a fearful succession of wild yells, an in discriminate massacre, and the demolition by fire of the habitations of their enemies, and then a hasty return with captives and bloody trophies of the pillage and butchery--these constituted, as a general rule, the sum total of a successful military excursion. " Their maner of warres," says Thomas Ilnriot,3 " amongst themselues is either by sudden surprising one an other most commonly about the clawnirig of the day, or moone light; or els by ambushes, or some suttle deuiscs: Set battels are very rare, except it fall out where there are many trees, -where eythcr part may haue some hope of defence, after the delruerie of euery arrow, in leaping behind.some or other." The Southern Indians are said
1 "Brevis Xarratio," plate xll.
2 Ibid., plate -\iv.
3 " A Briefe ami True Report of the New-found-land of Virginia," etc., p. '25.
Francoforti ad M cerium. DC Bry, anno 1590. Compare also Du Pratz's "His tory of Louisiana," chapter iii., book iv., sec. vil., vol. ii., p. 242, et serj. Lon
don, 1763. Smith's " History of Virginia," Richmond reprint, 1819, vol. i.,
p. 182.
18
ANTIQUITIES OF THE SOTJTIIEliN INDIANS.
to have dealt less cruelly with tlieir captives than did their T^ortheru neighbors. The Spaniards found to tlieir cost that tlie natives were very ready with tlieir weapons.
Says the Greutleman of Elvas,1 " Tlie Indians are so warlike and jiiml>le tLat they have no fear of footmen; for if these charge them, they flee, and when they turn their backs they are presently upon them. They avoid nothing more easily than the night of an arrow. They never remain relict, but are continually running;, trav ersing from place to place, so that neither cross-bow nor arcuiebuse. can be aimed at them. Before a Chris tian can make a single shot with either, an Indian will discharge three or four arrows, and he seldom misses of his object. AVhere the arrow meets with no armour, it pierces as deeply as the shaft from a cross-bow. Their bows are very perfect; the arrows are made of certain canes, like reeds, very heavy, and so stiff that one of them, when sharpened, will pass through a target. Some are pointed with the bone of a fish, sharp and like a chisel; others with some stone, like a point of diamond: of such, the greater nurabei1, where they strike upon armour, break at the place where the parts are put together; those of cane split, and will enter a shirt of mail, doing more injury than when armed."
A. public declaration of war was sometimes made by planting arrows along the pathway leading to the principal village of the enemy." They also were able, by means of ignited tufts of dried moss and grass, at tached' to the heads of their arrows, to set fire to thethatched cabins located in the fortified towns of their
WAULIKIi] CHARACTERISTICS. KIGII-PIIIEST.
10
adversaries. 1 The wretched crtielties visited even upon the dead bodies of the hostile slain are shock ingly portrayed byDeBry in plate xv. of the "Brevis N'arratio."
Courage, craft, perseverance, physical endurance, stoicism, ability to counsel with wisdom and.eloquence, experience In. combat, and activity and sliill in the use of "weapons, must all have been acquired and exhibited in a marked manner before the Southern Indian came to he regarded as a leading warrior in his nation. His honor and love of country lie esteemed of far greater value than life; and the most exquisite tortures failed to compel him to surrender and compromise the one, or renounce the other. In the arts of strategy, ambus"~*de, deception, and personal concealment, they excelled. -. i>j-A . Adair,2 in his general observations on the North American Indians, presents us with a detailed account of the martial spirit, devotion to country, caution In %var, method of fighting, cruelty to captives, fortitude in view of death, and the triumphs accorded to success ful warriors, as they existed among- the Southern In dians during the period of his residence among them, which, did the limits of this general sketch permit, we would gladly here reproduce.
A person of great power and consequence was the ANCEEKT HIGH-PRIEST. He presided iu spiritual af fairs ; and, In military matters, his influence was most potent. Never did the council determine upon a hos tile expedition without his counsel and sanction..3 To him was accorded the ability to hold personal com munion with invisible spirits capable of exerting a con-
20
ANTIQUITIES OF THE SOTTTIIETOT IKDIAS'S.
trolling influence over human plans. Through him the elements were propitiated, and his capability to foresee the result of an enterprise was unquestioned.
So great, remarks Mr. Uartram,1 is the influence of these high-priests, that they have been frequently known to-stop and turn back ail army after a march of several hundred miles, and when, it had approached within a clay's journey of the enemy-. Their predic tions were frequently surprising. They pretended, moreover1, to foretell the coming of a drought, and to be able to "bring rain upon the thirsty zea, to cure diseases, invoke or expel the presence of evil spirits, cause the tempest to cease, ami direct the thunder and lightning. It was their office to mediate between the beloved red-people and the bountiful, holy spirit of Fire. AVith their advice the season was set for plant ing, and occasions were designated for the solemniza tion of the public religious festivals. In every town they had their juniors or graduates learned iu the con jurer's and medicine-man's arts.
Among the Southern tribes the sun was regarded as the symbol of the power and beneficence of the Great Spirit, the Supreme God, or Creator, the soul and governor of the universe, the giver and takeraway of the breath of life. Hence, to this celestial luminary did they pay profound homage as to the visible minister and representative of the author of life, and light, and heat. To it their vows were offered as they puffed the smoke from the great calu met toward the heavens. "With reverence did they look upon the face of this God of Day, as they delib erated in council, or set out upon the war-path. Fire,
' Transactions of the American
SUN-WORSHIP. SUPERSTITIONS.
21
as an emanation from tliis celestial source, tliey vener ated and. propitiated with mysterious rites and cere monies. Temples were erected at great cost of mate rial and labor for this snn-worsliip, in which priests officiated. Their province it was to guard, tlie Eternal Fire in the Rotunda; and, in the solemn, annual festi val of tlie Busque, when all the iires of tile nation were extinguished, tlie high-priest alone--ministering between the Great Spirit and man--was commissioned, in the temple, to reproduce the celestial spark and give new fire to the community.
Believing in the immortality of the soul, in a future state of rewards and punishments, acknowledging the supreme power and control of one great, invisible, supreme spirit, these Southern Indians were plagued with ail apprehension of visions, dreams, trances, and malign influences of lesser divinities, which afforded ample scope for the operation by priests and conjurers using incantations, charms, and mysterious appliances upon their hopes and fears, credulity, and superstitions. TJpon the death of a high-priest, the entire community united in paying the fullest funeral -honors, and heaped above him the conical earth-mound. 1
If -we may credit the assertion of the Gentleman of Elvas," some of the Florida tribes worshipped the devil, and made offerings of human sacrifices to the spirit of evil.
Toward the latter part of February in each year, the Indians of Florida, taking the skin of the largest stag they had killed, stuffed it with the choicest fruits and matters which chiefly delighted them. The horns,
1 " Brevis Narratio," plate si. a " Narratives of the Career of ncrnando de Soto," etc., translated by Buck ingham Smith, p. .31. New York, 1806.
33
ANTIQUITIES OF TIIJE SOUTHER]? rs
neck, and. body ATere encircled with vines and fruits most Tare. Thus attired, the stag, with music and para.de, was carried and placed upon trie top of a tall tree, with its head and breast turned fall -toward the rising sun. By the king and high-priest--who stationed themselves nearest the tree--prayers 'were addressed to this celestial luminary, and petitions of fered that he would be pleased to reproduce the good gifts which were then presented. The members of the tribe assembled in a circle, and, at a little remove, re peated these supplications. When they were finished, all having saluted the sun, departed, leaving- the stuffed and garlanded stag until the recurrence of the same season, when, on each ensuing year, similar cere monies were observed.1
It is probable that some of the larger terraced mounds and truncated pyramids were temples erected in honor of, and devoted to the worship of the sun.
T\rithin the historic period idol-worsh ip existed, at least to a limited extent, among the Southern Indians. AVe will have occasion, however, in a subsequent chap ter, to consider this interesting- subject somewhat at length.
Among the Natchez the machinery of temples, idols, priests, keepers of sacred things, and sundry religions festivals, was most elaborate. The preserva. tion of the eternal fire engaged their utmost solicitude. The Sim ruled with despotic power, and seemed in his person to unite the privileges of ting and high-priest. Here were observed more emphatically than among any other Southern tribes the distinctions of ranlc The common people--or Miclie-Miche-Quipy (Stinkards) --were, to the last degree, submissive to the nobility,
3 " Brcris N" la-ratio," pfate ^xv.
THE STJK AMOJTG TJIE KATCHT:/.
23
consisting of Suns, nobles, and men of rank. These Suns claimed to be the descendants of the man and woman, who came down from the sun ; and their chil dren, to the remotest degree, were distinguished above the bulk of the nation and enjoyed an exemption from capital punishment. By them it was ordained that nobility should be transmitted only through the womei Upon the death of a Sun, many subjects. both m. and female, were sacrificed. No greater calamity could befall the nation than the extinction of the eternal iire. 1
The great chief of the Natchez bore the appella tion of TKE SiJ-N", He was succeeded in the kingly office by the son of the woman who -was most nearly related to him.2 To this woman the title of woman chief. was given. Great honors were paid to her, al though she meddled not in affairs of state. Like the great chief, she possessed the power of life and death over the common people, and did not hesitate to order her guards to slay any who offended her.
Every morning, says Father Charlevoix,* as soon as the sun appears, the great chief comes to the door of his cabin, turns himself to the east, and howls three times, bowing down to the earth. Then they bring him a calumet, which is used only for this pur pose. This he smokes, and blows the smoke of the to bacco first toward the sun, and then toward the other three cardinal points. He acknowledges no superiorother than the sun. From this luminary he claims to
1 Sec D;i Pratz's " History of Louisiana," TM1. ii., chap. Hi., sec. 2-4, pp. 170, 222. London, 1763.
j4.raong tlje Carolina Indians the succession fell not to thy king's his sisU'r's son (Lawson's " History of Carolina," p. 195. London, it appears that a similar rule obtained among other Southern tribes.
3 '- Voyage to Xorth America," etc., rol. ii., p. 100. Dublin, 17t>6.
24
ANTIQUITIES OF THE SOTJTIIEEW INDIANS.
derive his origin. Over his subjects lie exercises un limited power, can dispose at pleasure of their prop erty and lives, and pays no recompense for any labors he may demand of them.
The deatli of a great chief costs the lives of his guards, and sometimes of more than a hundred per sons. At one time very few of the principal personages died without being escorted to the country of souls by some of their relations, friends, or servants. Suspect ing the death of De Soto, the Cacique of Guachoya ordered two well-proportioned young Indian men to be brought, saying it was the usage of the country, when any lord died, to kill some persons who might accom pany and serve him on the way to the spirit-land. He, therefore, ordered their heads to be struck off, and it was only after much persuasion, and upon the em phatic statement that the governor was not.dead but had only gone on a visit to the heavens, attended by a suitable number of soldiers, that Luys de ]Moscoso succeeded in effecting the release of these young Indians. 1
The tribes encountered by De Soto during his march east of the Mississippi were ruled over by caciques to whom their subjects yielded implicit obe dience. The province of Cutifachiqui, however, was governed by a eacica who welcomed the Spanish ad venturer right royally, and extended to him the hos pitalities of her kingdom. The stern of her canoe was covered with an awning, and she sat upon cushions. The country was delightful and fertile, arid here were found, in the possession of the natives and in the barbacoas, large quantities of clothing, and shawls
1 "Narratives of the Career of Hernando De Soto," translated by BuckingHam Smith, p. 148. New York, 1866.
MAUSOLEUM AT TA1OMECO.
25
made of fibres of the "bark of trees and of feathers richly colored in white, gray, vermilion, and yellow, suitable for winter wear--well-dressed deer-skins with various designs depicted upon them--and many pearls. The inhabitants, says the Gentleman of Elx'as, -were brown of skin, well formed and admirably propor tioned. He distinctly affirms that they were more civilized than any peoples ho had seen in all the terri tories of .Florida, and that they wore clothes and shoes. To this cacica her subjects paid great respect, and her niece was at first commissioned to meet De Soto and assure him of the good-will of the queen. Mention is also made of the queen-mother, a widow, who, repos ing upon her dignity, refused to hold converse with the strangers.
At Talomeco was a mausoleum a hundred paces in length and forty in breadth, with lofty roofs of reed. The entrance to this temple was guarded by gigantic wooden statues, carved with considerable skill, the largest of them being twelve feet high. Armed with various weapons, they stood in threatening attitudes and -with ferocious loots. Within "were statues of various shapes and sizes. Around the sepulchre were benches upon which, in "wooden chests skilfully wrought, but without locks or hinges, reposed the bodies of the departed caciques, priests, and chieftains of Catifachicmi. Beside these were smaller chests, and cane baskets filled with valuable furs, robes of dressed skins, and mantles made of the inner rind of trees and of a species of grass which, when beaten, closely resem bled flax. There were coverings formed of feathers of various colors, which the natives wore in winter. This temple also contained great store of pearls.
Adjacent to this grand sepulchral receptacle were
26
ANTIQUITIES OF THE SOUTHERN INDIANS.
several buildings which served as armories. In. them weapons of various sorts were carefully arranged and kept in order by numerous attendants.' The erection of temples or "buildings, and their dedication to the preservation of the bodies of their chiefs obtained also among other Southern tribes. Thus, in plate xxii. of the " Admirarula Narratio," we have a representa tion of one of these sepulchres as it existed among the Virginia Indians in the sixteenth century. The ex planatory note is thus quaintly " made in English " by Thomas Harlot, servant to Sir "Walter lialeigh :
THE TOilDE OE THEIK WEttOWAXS OK dlEIFF LORDES.
" Tlie bnilcle a Scaffolde 9 or 10 foote hihe as is expressed in this figure under the tobs of theit AVeroans, or cheefe lordes which they couer with matts, and lai the dead corpses of their weroaiis thereuppon in manner followinge. First the bowells are taken forthe. Then layinge down the skinne, they cutt all the flesh cleaiie from the bones, which the drye in the soniie, and v\rell dryed the inclose in Matts and place at their feete. Then their bones (remaininge still fastened together with the ligaments whole and vncorrupted) are couered agayne with leather, and their carcase fashioned as yf their nesh "wear not taken, away. They lapp eache coi'ps in his owiie skiiine after the same in thus handled, and lay yt in his order by the corpses of the other cheef lordes. By the dead bodies they sett their idol Kiwasa, whereof we spake in the former chapiter. For they are per suaded, that the same doth kepe the dead bodyes of their cheefe lordes that nothiiige may hurt them. Moreoiier under the foresaid scaffolde some 011 of their
1 Ivvin's " Conquest of Florida," chapter xlviii.
TOMBS OF TI-Uj VIUCflKTA
preists liatli his lotlginge. which Miimbleth his prayers iiighte and day, and liatli charge of the corpses. For his bedd lie hath two deares skinnes spredd on the grotvnde, yf the -wether bee cold hee maketli a fyre to warine by \vithall. Thes poore soules are thus iiistructed by natute to reuerouce their princes enen after their death." 1
Caves were sometimes dedicated to similar uses.
1 " ABriefe and True Keporfc of the in English by Thomas Harlot." Plate xs Mcenum. Do Bry. Anno 1590.
CHAPTER II.
Office of the Conjurer or Medicine-man.--Treatment of the Sick.--Medicinal Plants.--Towns and Private Houses.--Tenure of Property.--Agricultural Pursuits.--Town Plantations and Private Gal-dens.--Public Granaries.--Ani mal and Vegetable Pood.--Mechanical Labors.--Early Mining in Duke's-Creek Valley.--Manufacture of Canoes, Pottery, Copper Implements, Gold, Silver, shell, and Stone Ornaments.--Various Implements and Articles of Stone, Bone, and Wood.--Trade Relations.
ASOTIIEE important person in every community was the CO^JTJREK, who generally united in himself tlie offices of priest, physician, and fortune-teller. He was supposed to possess unusual powers "because of his constant communion with and influence over evil spir its. Various and extravagant were his incantations, his charms mysterious and unexplained, and his con tortions, when engaged in the practice of his arts, pro longed and violent. His knowledge of medicinal herbs and simples gave him a decided advantage over the unlearned. Encouraging the superstitions of his patients, he pretended to work wonderful cures, and acquired wherever he went an influence most marked in its character. In the explanatory note accompany ing' plate ss. of the " Brevis ISTarratio," we are made acquainted with several methods adopted by the Flor ida Indians in their treatment of the sick. One rem edy consisted in scarifying the forehead of the patient with a, shell, and sucking therefrom the blood and hu-
TJREATMENT O-F THE SICK. JAOUSAS.
29
inors which were supposed to contain the seeds of tlie disease. Others, suffering from different maladies, were compelled to lie upon their stomachs with their heads over pans, from which they inhaled, through their mouths and nostrils, the fumes of certain medici nal plants in a state of ignition. Tobacco-smoking was also employed as a means of expelling disease. To Coreal * we are indebted for the . following inter esting account of the office of the medicine-men among the Florida tribes:
" AVhen they are siclc they have not a vein opened, according to our practice, but send for their Jaounaa who are their priests and physicians. The latter suck that part of the body which causes the patient the greatest pain, and this they do with the mouth, and sometimes also by means of a kind of shepherd's flute (une espece de chalumeau), after having made a small incisioii near some vein. They also mate incisions in the suffering parts of those who submit to their treatment. Previous to the ceremony, and also after the operation, the jaoiina utters some words. "Whether the patient dies or recovers, the jaoi'ina's gravity re mains unaffected. This behavior constitutes a part of his professional art. The respect and confidence with which the savages regard these men remain the same, no matter what the result may be.
"The jaotinas also understand how to make their patients vomit by means of a powder which they pre pare from calcined shells. One must be a Floridian or the devil to resist the violence of this emetic, for I doubt whether there exists a more efficient prescrip tion for sending a European to the other world. They
''Voyages de FraDcois Coreal, aux Indes Occidentals (1600-169*7)." Am sterdam, 1*722. Vol. L, pp. 3y-41.
30
ANTIQUITIES OF THE SOTTIIKEN INDIAKS.
also bathe their sick, and when all remedies are ex hausted, and no hope remains, they expose them before their cabins to the rising sun, imploring and conjuring that luminary to cure them. In all diseases the meth od of treatment (or succession of remedies) is the same. They begin by sucking and making incisions, after which they resort to bathing, etc., until recovery or death occurs. In all these proceedings they preserve vrell their presumption, which they conceal from these poor people under an affected modesty and feigned abstinence. It is true, however, that they go through a rough and long apprenticeship under the Q\<\. yaounas, who are the chiefs of the sect, and this doubtless en hances the confidence which the Floridians repose in these priests and physicians. These jaoiinas are clacl in long robes made of various skins cut into unequal bands. These robes are fastened by girdles of deer skin, to which they attach their pouches tilled with herbs. Over the robe they wear, after the fashion of a cloak, the hide of some wihl animal. Their feet and arms are uncovered, and they have on their heads a skin cap terminating in a point."
Of the Virginia conjurers, Hariot' says, they use strange gestures and are often contrary to Nature in their enchantments. They be " verye familiar with deuils, of whom they euquier what their enemys doe, or other suche thinges. They shaue all their heads sauinge their creste, which they weare as other doe, and fasten a small black birde aboue one of their ears as a badge of their office. They weare nothinge but a skinue, which haugeth downe from their gyrdle and couereth their prinityes. They weare a bagg by their
PHYSICIANS AND CONJUKIEBS.
31
side. The Inhabitants giue great credit viito their speeche, which oftentimes they finde to bee true."
The Natchez jugglers not only pretended to cure the sick, but also professed to procure rain and sea sons favorable for tlie fruits of the earth. Their in cantations were often directed to the dispersion of clouds and the expulsion of evil spirits from tlie bod ies of the afflicted. They were a lazy set of fellows, imposing trpon. the credulity of their countrymen, and receiving rich rewards when their patients recovered. 1 The Alibamons reposed great confidence in their doc tors, and regarded the ravings of these quacks and cunning impostors as the utterances of a divine lan guage."
Among the Carolina tribes the priests were the 1 conjurers and doctors of the nation.' The theory was
that all distempers were caused by evil spirits; conse quently, none of their physicians attempted to effect a cure until he had conversed with the good spirit, and ascertained wh ether his aid. could be-secured in the effort to exorcise the adverse demon.
" As soon as the Doctor comes into the Cabin," says Surveyor-General Lawsoii,* " the sick Person is sat on a Mat or Skin, stark-naked, lying on his Back, and all uncovered, except some small Trifle that covers their Nakedness when ripe, otherwise in very young Children; there is nothing about them. Ill this Manner the Patient lies, when the Conjurer appears ; and the King of that ISTation comes to attend him with a Itattle made
1 See CbarlcYoix's "Voyage to North Americn," etc., vol. ii., p. 203. 11611.
2 "Travels through Louisiana, by Captain Bossu," yol. i., p. 2G4. 17*1.
3 Lavrsoivs " History of Carolina," p. 211. London, 1711. 4 Idem, p. 214.
Dublin, London.
32
ANTIQUITIES OF THE SOTJTIIEBN INDIANS.
of a Gourd with Pease in it. This the King delivers into the Doctor's Hand, whilst another "brings a Bowl of "Water, and sets it down. Then the Doctor begins, and utters some few Words very softly ; afterward s he smells of the Patient's Navel and Belly, and sometimes scarifies him a little with a Flint, or an Instrument made of Kattle-Snake's teeth for that Purpose ; then he sucks the Patient, and gets out a Mouthful of Blood and Serum, but Serum, chiefly ; which, perhaps, may Tje a better Method in many Cases, than to take away great Quantities of Blood, as is commonly practised ; which he spits in the Bowl of "Water. Then he begins to mutter and tall?: apace, and, at last, to cut Capers and clap his Hands on his Breech and Sides, till he gets into a Sweat, so that a Stranger would think he was running mad; now and then sucking the Patient, and so, at times, keeps sucking, till he has got a great Quantity of very ill-coloured Matter out of the Belly, Arms, Breast, Forehead, Temples, Neck, and most Parts, still continuing his Grimaces and an tick Postures, which are not to be matched in Bedlam. At last you will see the Doctor all over of a dropping Sweat, and scarce able to utter one Word, having quite spent himself; then he will cease for a while, and so begin again till lie comes in the same pitch_ of Having and. seeming Madness as before. (All this time the sick Body never so much as moves, although, doubtless, the Lancing-aud Sucking must be a great Punishment to them ; but they cer tainly are the patientest and most steady People under any Burden that I ever saw in my Life.) At last the Conjurer makes an end, and tells the Patient's Friends whether the Person will live or die ; and then one that waits at this Ceremony takes the Blood away (which remains in a Lump in the middle of the Water),
MEDICINE--MEN.
83
and. buries it in the Ground, in a Place unknown to any one but he that Intel's it." ' " In Medicine, or the Nature of Simples," says Thomas Ash,* " some have an exquisite knowledge ; and in the Cure of Scorlrutic, Venereal, and jMaliynant Distempers, are admirable. In all External Diseases they suck the Fart affected, with many Incantations, 1'hiltres, and C'7tarms."
These medicine-men also conjured for stolen goods, understood the art of coloring the human hail", cured lingering distempers by wrapping a snake around the body of the afflicted, treated affections of the spleen and of the stomach by hot applications, relieved the toothache, administered ample purges through large draughts of the Yizupon, comprehended the medicinal virtues of tho sassafras and. many native plants, ap proved of the salutary influences of profuse sweat ing, rubbed with the fat of animals to render the limbs pliable, and, when wearied, to relieve pains iu the joints, administered the juice of the tulip-tree as a remedy for pox, and suggested various specifics for dis eases incident to climate and the exposed manner of life."
The office of physician among these primitive peo ples, accompanied as it was with authority, notoriety, and emolument, was not exempt from danger. Fail ure to effect a cure, in some instances, involved, as a direct con sequence the death of the practitioner. The suggestion of such a penalty at this time, for profes sional ignorance or malpractice, would most essentially diminish the applications for admission to the degree
1 Compare Brickell's " Xntur.il History of North Carolina," p. 072. 1767.
* "OaroMan," etc., p. 86. London, 16S2. " Stc Lawson's " Carolina," p. 215, rf M-?. London, 1714
DuUin,
34
ANTIQUITIES OF THE SOUTHERN INDIANS.
of M. D., emphatically tliiu the ranks of the medical fraternity, and entirely extirpate the race of imptident quacks infesting the country through all its bor ders.
In the opinion of Mr. Bartram,* the Southern In dians, in the treatment of diseases, depended more upon regimen and abstinence than they did upon, medi cines.
The Cherokees used the Lobelia syphilitic, and endeavored to conceal from the whites all knowledge both of its virtues and of the localities in which it grew. A decoction of the Hignonia crucigera and of the roots of the cliina brier and the sassafras was freely employed for the purification of the blood. The caustic and detergent properties of the roots of the white nettle (rTcttroplia itrens) were utilized in cleansing old nlcers and. consuming proud flesh, while the dissolvent and dinretant powers of the root of the Convolvulus panduI'atus were highly esteemed as a remedy in nephritic complaints. The emollient and discutient power of the swamp-lily (tSaururus cernuus') and the virtues of the Jiypo or may-apple (Podophyllujn peltat'imi~) were both communicated to the Europeans by the Indians.
The roots of the Panax ginseng and J^orida, or white-root, were held in the highest esteem among the Cherokees and Creeks. The virtues of the former are well known, and the friendly carminative qualities of the latter were constantly invoked for relieving all dis orders of the stomach and intestines. The patient chewed the root and swallowed the juice, or smoked it, when dry, with tobacco. Even the smell of the root exerted a beneficial effect. The Lower Creeks, in whose
J
ANCIENT TOWNS OF FLORIDA.
35
country it did not grow, gladly exchanged two or three buckskins for a single root of it.*
Of tlie ANCIENT TOWNS of Florida, De Bry has given us several representations." They are all small, circular in outline, and defended "by stockades. The dwellings of chiefs, council-houses, public "buildings, granaries, and temples, we have considered elsewhere, and it only remains for us, in this connection, to notice the charac ter of the cabins occupied by the common people. These were confined, inconvenient, and ephemeral in their structure. Describing the dwellings of Toalli, the Knight of Elvas remarks that they were roofed with cane after the fashion of tile. They were kept very clean, and their sides, made of clay, looked like tapia. Throughout the cold country, he continues, every Indian has a winter house, plastered inside and out, with a very small door, which is closed at dark. Within, a fire is kindled which heats the building like an oven and renders clothing during the night-time en tirely unnecessary. The summer-house was more open, and near it was erected a small kitchen for baking bread. Maize "was kept in a house with wooden sides, raised aloft on four posts, -with a cane floor. The houses of the principal men or chiefs were larger than those of the subjects, and, in front, had deep balconies furnished "with cane seats. There were also large barbacoas filled with maize, deer-skins, and the blankets of the country--the tribute of the common people to their rulers.' These private residences were generally
1 Consult also-Adults "History of the North American Indians," p. 172, ci seq. London, 17M.
2 "Brevis Narratio," etc., plates xxx., xxxi., xxxiii., si. Francoforti ad Mcenum, anno 1591.
s See " Narratives of the Career of Hernando de goto," translated by Buck ingham Smith, p. 52. New York, 1800,
36
ANTIQUITIES OF THE SOUTIIEP.X INDIANS.
circular in form, tlieir sides made of upright poles, the interstices filled witli clay, and tlieir tops thatched with dry grass, palmetto-leaves, or cane. During the sum mer months but little covering was needed," and the light, open summer-houses were frequently roofer! sim ply with tranches of trees. At a later period the Muscogulgees built houses much more substantial in tlieir character, plastering the walls carefully with red or white clay, and ornamenting them with various draw ings of animals, plants, trees, birds, and ,111611.'
In the latter part of the sixteenth century the houses of the Virginia Indians were made of poles fastened at the top and covered either with bark or with rush mats. They were from twelve to twentyfour yards in length, and. about half as broad. Their towns consisted of a collection of from ten to thirty houses, and were sometimes open, and, in other in stances, were protected by stockades not unlike those in use among the Florida tribes. An example of a for tified village is presented in plate xix. of the " Adrniranda Karratio." The town of Secota (plate xx.), on the contrary, is entirely unprotected. In the vicin ity of this village are seen fields of maize and tobacco. The relative positions of the places of prayer, of feast ing, of dancing, of idol-worship, of the spot where the sacred, fire is kept burning, of the large building wherein, are entombed their kings, and the locality whence they derived their supply of water, are all delineated. AVhen a village was situated at a remove from a stream, spring, or lake, the earliest attention was paid to digging an artificial pond from which a liberal supply of water could at all times be obtained.
3 Bartram's " Observations on the Ci-cek and. Cherokce Indians." Transactions of the American Ethnological Society, vol. iii., part 1, p. 18.
WIGWAMS OF THE CAEOT-I^TA I^.'XHAX"S.
ot
Bibault thus describes a native village on the Flor ida coast: " Their houses be made of wood fitly and close; set upright and covered with reed s, the most part of them after the fashion of a pavilion. But there was one house among the rest very long and broad, with settles about made of reeds, trimly couched together, which serve them both for beds and seats ; they be of height two foot from the ground, set upon great round pillars painted with red, yellow', and bine, well and trimly polished." '
Perhaps the most minute and satisfactory descrip tion of the dwellings of the Southern Indians is that presented by Mr. Ijawson.2 Referring more particu larly to the Carolina tribes, he writes: L' These Savages live in Wigwams or Cabins built of Bart, which are made round like an Oven, to prevent any Damage by hard Gales of "Wind. They mate the Fire in the mid dle of the Ffotise, and have a Hole at the Top of the Hoof right above the Fire, to let out the Smote. These Dwellings are as hot as Stoves, where the Indians sleep and sweat all ISTight. The Floors thereof are never paved nor swept, so that they have always a loose Earth on them. They are often troubled "with a multitude of Fleas, especially near the Places where they dress their Deer-Stins, because that Hair harbors them; yet I never felt any ill, unsavory Smell in their Cabins, whereas, should we live in our Houses, as they do, we should be poison'd with our own hastiness ; which confirms these Indians to be, as they really are, some of the sweetest People in the ^Vorld.
" The Bart they mate their Cabins withal, is gen erally Cypress, or red or white Cedar; and sometimes,
38
ANTIQUITIES OF THE SOtirHERS UTDIAITS.
when, they are a great way from any of these \Voods, they make use of Pine-Bark, which is the worser sort. In building these Fabricks they get very long Poles of Pine, Cedar, Hiccory, or any AVood that WJH bend ; these are the Thickness of the Small of a Man's Leg at the thickest end, which they generally strip of the Bark, and warm them well in the Fire, -which makes them tough and fit to Lend ; afterwards they stick the thickest ends of them in the Ground, about two Yards asunder in a Circular Form, the Distance they design the Cabin to be (which is not always round, but some times oval), then they bend, the Tops and bring them together, and bind their ends with. Bark of Trees, that is proper for that use, as Elm is, or sometimes the Moss that grows on Trees, and is a Yard or two long, and never rots; then they brace them with other Poles to make them strong; afterwards cover them all over with Bark, so that they are very warm and tight, and "will keep firm against all the "Weathers that blow. They have other sorts of Cabins without Windows, which are for their Granaries, Skins, and Merchandizes; and others that are covered over head ; the rest left open for the Air. These have Reed-Hurdles like Ta bles to lie and sit on, in Summer, and serve for pleas ant Banqueting-Houses in the hot Season of the Year. The Cabins they dwell in have Benches all round, ex cept where the Door stands ; on these they lay BeastsSkins and Mats made of Rushes, whereon they sleep and loll. In one of these several Families commonly live, though all related to one another." *
CABINS OF THE GEOKGIA ISDIAXS.
3!)
Some fifty yeai'S subsequent to the time when, tliis description was furnished by Mr. Lawson, De Bralim thus perpetuated his observations respecting the char acter of the houses used by the Indians on the coast of South Carolina and Georgia : " The Indian built their houses of posts, on which they lash in and out side canes, and plaster them over with a white clay mixed with small pieces of talk, which, in a sun shiny day gives to these houses, or rather cottages, a splendor of unpolished silver: they are about twelve foot wide, and twenty or more foot long, covered "with a clap board roof, have no windows, but two doors on the opposite sides, sometimes only one door ; the fire place is at one end of the house, with two bed states on both sides of the fire ; the bed states are made of canes, raised from the ground about two foot, and covered with bear's skins ; their corn houses are buit in the same manner, but raised upon four posts, four and five foot high from the ground; its floor is made of round, poles on which the corn worms cannot lodge, but fall through, and thus the Indians preserve their corn from being distroyed by the weevils a whole year. Two or more famelies joine together in building a hot house about thirty foot in diameter, and fifteen foot high, in a form of a cone, with poles and tatched, without any air hole, except a small door about three foot high and eighteen inches wide ; in the center of the hot-house they bum fire of well seasoned dry wood; 'round the inside are "bedstades fixed to the studs which support the middle of each post; in these houses they resort with their children in the winter nights. Upon the same plan of these hot-houses (only a greater diameter and perpendicul) their town houses are built, in which the head men assemble to consult
40
ANTIQUITIES OF THE SOUTTLEEiX INDIANS.
in war, peace, or other concerns and. every evening1 during summer all families of the town meet to dance and divert themselves." 1
In the vicinity of the Georgia villages, which, were usually located upon the banks of streams and in rich valleys, the followers of De Soto often found, large fields of maize, and in some instances artificial lakes used, as fish-preserves.
With reference to the TJSXTTRE OF T.AZSTD AND PROP ERTY, it may be remarked, generally, that every tribe had its boundary-lines, and each nation or confederacy its own recognized territorial limits. In the public domain, with its rivers, and lakes, and forests, each Indian claimed a right of property for the purposes of travel, hunting, and fishing. All that a man earned or fashioned by his individual labor and industry be longed to himself, and he could dispose of it accord ing to the customs and usages of his people. It -was his privilege to clear, settle, and plant as much. land. as he chose, within the boundaries of his tribe.
In villages the right of personal property was scrupulously observed, and theft was an uncommon occurrence. Every town or community, for the sake of convenience, assigned a parcel of land in its vicinity for agricultural purposes.
This was called the " town plantation,2 where every family or citizen had his parcel, or lot, according to desire, or convenience, or the largeness of his family." These shares "were bounded by a strip of grass, by poles, or some artificial marks. In ancient
AGEICTJLTTJHAI, 1ABOKS.
41
times, iii these corn-fields there were no fences. Eacli person, however, recognized the limits of his own. little farm, and refrained from interfering with Ms neigh bors' rights. 1 The entire plantation, therefore, was simply a collection of lots, adjacent the one to the other, and all embraced in one general enclosure. AVhen the proper season for planting arrived, all the inhabitants, as one family, devoted, their attention to the preparation, of the ground and the sowing of the seed. In like manner the plants, at proper times, "were, by common consent, cultivated. These agricul tural labors were superintended by an overseer elected or designated annually for that purpose. During the periods of special labor his province it was to awaken, the inhabitants of the town at daybreak -with a singu larly loud cry, assemble them with their agricultural implements in the public square, and, by sunrise, lead them into the fields where the work was commenced, and under his supervision prosecuted until evening. The women did not march out with the men, but .fol lowed in detached parties bearing the provisions of the day. " AVhen the fruits of their labors are ripe and in fit order to gather in," says Mr. Bartram, " they all, on the same day, repair to the plantation; each gathers the produce of his own proper lot, brings it to town, and deposits it in his own. crib, allotting a cer tain portion for the public granary, which is called the king's crib, because its contents are at his disposal, though not his private property, but considered as the tribute or free contribution of the citizens of the state, at the disposal of the king.
" The design of the common granary is for the wisest and best of purposes with respect to their people, i. e.,
1 "Luwson's Carolina," p. 179. London, 1714.
42
ANTIQUITIES Of THE SOUTHERN INDIANS.
a store or resource to repair to in cases of necessity. Thus, when a family's private stores fall short, in cases of accident or otherwise, they are entitled to as sistance and supply from the public granary, lay ap plying to the king. It also serves to aid other towns which may be in want, and affords provisions for their armies, for travellers, sojourners, etc., etc. Thus the inico becomes the provider or father of Ms people."
Besides the general plantation, each inhabitant of the village enclosed a garden-spot adjoining his cabin, in which he cultivated corn and vegetables, upon which he subsisted before the general harvest was
gathered. "Widows with largo families were always assisted
in planting, working, and gathering their crops.' Throughout the Creek Confederacy there wras con
tinual and friendly intercourse between the families constituting the respective tribes. To their doors there were no bolts, and universal hospitality and good feeling prevailed. Ever ready to assist each other, and entertaining an abiding friendship, the one for the other, the members of the various tribes seemed to Mr. Bartram to constitute one great family, holding all their possessions in common. Theft was almost
unknown. The AXIMAL POOD of the Southern Indians, at the
dawn of the historic period, comprised all the wild animals native to the region, among which may be specially enumerated buffaloes, deer, bears, beavers, panthers, raccoons, opossums, wild-cats, rabbits and
ASTJIAL FOOD.
43
squirrels. These wore generally killed with the bow and arrow. Certain seasons of the year were set apart for hunting, during which large quantities of meat were obtained, cured, and housed, for future consump tion. Fawns in the womb were esteemed a great deli cacy. All sorts of fishes, turtles, terrapins, oysters, clams, fresh-water mussels, conchs, alligators, arid even some varieties of snakes, were eaten, and much time was consumed in the capture of fishes by means of the bow and arrow, spears, nets, baskets, and "wears. The bone hook, and line made of deer-thong, or twisted fibro, were used only to a limited extent. Captain John Smith 1 asserts that the Virginia Indian women spun betwixt their- hands and thighs the barks of trees, a kind of grass, and deer sinews, out of which they readVily made a very even thread. Out of this thread they made garments, nets, and fishing-lines. "Their fish hooks," he continues, " are either a bone grated as they noch their arrowes, in the forme of a crooked pimie, or of the splinter of a bone tyed to the clift of a little sticke, and with the end of the line they tie on. the bate." Young wasps, "white in the comb, were re garded as a dainty morsel. Wild-turkeys, water fowl, and various birds, were eagerly sought after and eaten. In a word, there was but little animal life in the forests or in the "waters of the country which the Southern Indian excluded from, his food-list. Even
1 "History of Virginia," Richmond reprint, 1819, vol. i., p. 133. 2 Lawson's " History of Carolina," p. 178. London, 1714. "Brevi3 Narratio," plates xxiv., xxv., xxvi. Francoforti ad Moauum, 1591. " Admiranda Narratio," plates siii., xiv. Franeoforti ad Mrenum, 1590. Ash's " Caro lina," p. 36. London, 1682. Bartram's " Observations on the Creek and Cherokee Indians." Transactions of the American Ethnological Society, vol. iii., part i., p. 47. Adair's " History of North American Indians," p. 402, et sty. London, 1775. Timbcrlake's "Memoirs," p. 45. London, 1765. Smith's "History of Virginia," Richmond reprint, 1819, vol. i., p. 133.
44
ANTIQUITIES OF TI-IE SOCTirERX IS"DIAK"S.
upon dogs did they sometimes subsist. The skills of the buffalo, deer, and other animals, were dressed and used as clothing.
Among the vegetables upon which these primitive peoples chiefly relied for sustenance, may be mentioned Indian corn (maize or zea), wild-potatoes, ground-nuts, acorns, walnuts, hickory-nuts, chestnuts, pumpkins, melons, gourds, beans, pulse of various sorts, persim mons, peaches, plums, grapes, and mulberries. The tuberous roots of the sm.ilax (/S. pseudochinci^) were dug up, and, while still fresh and full of juice, were chopped up and macerated well in wooden mortars. "When thoroughly beaten, this pulpy mass was put in earthen vessels containing clean water. Here it was stirred with wooden paddles or "with the hands. The lighter particles, -floating upon the top, were poured off. A farinaceous matter was leffc at the bot tom of the vessel; which, when taken out and dried, remained an impalpable powder or farina of a reddish color. Boiled in water, this powder formed a beauti ful jelly, which, when sweetened, was both agreeable and nourishing. In combination with corn-flour and when, fried in fresh bear's-grcase it made excellent fritters. 1
Tobacco also was regularly and extensively culti vated. The Southern Indians, especially those resident upon the rich valleys of the interior, devoted no little time and attention to agriculture. With them maize was emphatically the staff of life. Upon its nutritious properties they relied both during its milky state and when dry. In the latter condition it was often parched, pounded, moistened with water, and thus eaten. This
1 Bar tram's " Observations on the Creek and die of tljc American Ethnological Society, vol. iii., part
USE OF CORIS", ircne, AND SALT.
40
was the case when the party was on the march or en gaged in hunting. Greiierally "beaten in a mortar, it was eitlier boiled for hominy, or, mixed with hickorynut-milk, walntit-oil, or fresh bear's-fafc, was baked into bread or fried, as cakes. In a subsequent chapter upon agriculture we "will note more carefully the facts con nected with the cultivation, preservation, and use of the grain which subserved such important purpose in the domestic economy of these peoples.
"Walrmts arid hickory-nuts were dilig-ently collected, cracked, and boiled in vessels, when the oil which rose to the surface was skimmed off and carefully preserved in covered earthen jars. This oil was highly esteemed. in the preparation of their corn-cakes. Of the seeds of the sunflower, when, pounded, they also made bread. The amexlas was freely eaten, and. ripe persimmons were pressed into cakes and stored away for consump tion during the winter months. Grapes were dried in the sun an d collected i n the public granaries and private store-houses. Wild-honey was also gathered. 1
Salt was manufactured by the natives. The Knight of Elvas 2 informs us that the natural salt uud the sand with, which, it was intermixed were thrown, into bas kets made for the purpose. These were large at the mouth and small at the bottom, or, in other words, fun nel-shaped. Beneath them--suspended in the air on a ridge-pole--vessels were placed. "Water was then poured upon the admixture of sand and salt. The
1 Consult " Narratives of the Career of Hernamlo de Soto," etc., translated by Buckingham Smith, pp.38, 55, 69, 77, 200-202. New York, 1S6G. " A Briefe and True Report of the New-found Land of Virginia," hy Thomas Hariot, pp. 13-16. Francoforti ad Mcemiin, 1590. Bossu's "Travels through Louisiana," vol. i., p. 224. LawFon's "History of Carolina," p. 207. London, 1714. "Brevis
s " Narratives of the Career of Hernando de Solo,'' etc., p. 124. Xe\v York, 1866.
46
ANTIQUITIES OF THE SOUTHERN INDIANS.
drippings were strained and foiled 011 tlie fire until all the water was evaporated, and trie salt left in the bot tom of trie pots.
Captain Komans asserts tliat the Indians never ate salt meats or foiled their food with salt. Nevertheless they had salt in abundance. ^Vlien deprived of it for a long time, he says an Indian " will frequently eat a pound of salt without anything else." 1 To the saline springs of Tennessee and Kentucky the natives constantly resorted from time immemorial, arid in large numbers, for the manufacture of this necessary season ing for food. They also obtained rock-salt from nat ural deposits near the mouth of the Mississippi Hiver.
Of the MECHANICAL IABOKS of the aborigines--aside from, the construction of their tumuli, fortifications, fish-preserves, temples, public and private liouses, and places for feasting, dancing1, and religious exercises --it may be remarked that, in the manufacture of pottery, from its most careless expression in small terra-cotta pans, or gourd-shaped drin"king-cups, to its more substantial development in burial-vases, large, ornamented cooking-vessels and well-formed jars for the preservation of fruits and oils, the Southern Indi ans excelled. They had made fni'ther progress in the ceramic art than that attained by the ^Western and Northern tribes. Their pottery savored less of the archaic type, and ill form and ornamentation, as well as in smoothness and homogeiieousiiess of composition, gave evidence of superior taste and skill. The shapes of these fictile wares were also more varied. They understood and practised the art of mixing their well kneaded clay with pounded shells and gravel, so as to impart to the material greater tenacity and dura-
ren- York,
POTTERY, COPPER IMPLEMENTS, ETC.
47
bility. Tlie ornamentation, of the rims, necks, and. sides of this earthenware was varied, and often tasteful and ingenious. The use of tlie potter's "vvlieel seems to have been unknown. To tlie women was chiefly committed the manufacture of this pottery. Soapstone, in many localities, was the favorite material from which, "by means of flint implements, were fash ioned culinary utensils, both great and small. jSTo implements of iron and bronze existed at this early period, and. copper was used only to a limited extent. In its treatment that material was regarded rather in the light of a malleable stone, than as a metal. Its employment was confined almost exclusively to the manufacture of ornamental axes, goi'gets, pendants, ancl spindles, or points for piercing pearls. Procured, in a pure, native state--chiefly from the shores of Lake Superior--it was, while cold, hammered out into the desired shape. Heat was never applied, and all the implements and ornaments of this metal, which we have seen, show very plainly a laminated struutcre. Comparatively few copper articles have been found within the limits of Georgia, and most of these, as we shall hereafter observe, were obtained from ancient graves in the valleys of the Chattahoochee, the Etowah, and the Oostenaula.
Gold and silver, to a limited extent, were employed in the fabrication of ornaments. Small masses of these precious metals were picked up by the natives in pockets, or gathered in the beds of streams flowing through atiriferous regions, aud perforated and worn as pendants. Gold beads--evidently not European in their manufacture--rudely hammered into round and oval shapes, with holes drilled through their centres or upper portions, have been, found in the Etowah
43
ANTIQUITIES OF THE SOUTHERN
Valley, in the vicinity of the large mounds oil Colonel Tuinlin's plantation. In this connection, it is proper that we allude to the traces of early mining in Cherokee Georgia.
In 1834, Colonels Merriwether and Lumsdeu, while engaged hi digging a canal ill Duke's-Creek Valley for the purpose of facilitating their mining operations, imcarthed a subterranean village consisting of thirty-four small cabins, located in a straight line extending up ward of three hundred feet. They were made of logs hewn at the ends and notched down, after the fashion of the rude log-huts of the present day. This hewing and notching had evidently been, done with sharp metallic tools, the marks being such as would hare been caused by a' chopping axe. Above these little houses--situated from fifty to one hundred yards from the principal channel of the creek, and embedded from seven to nine feet below the surface of the ground-- trees were growing from two to three feet in diameter. 1 The estimated age of these trees was somewhat over two hundred years. The violent changes often caused, ill their narrow valleys and along their yielding banks, by mountain-streams swollen "with rain or engorged by the dissolving snows of winter, may account for the inhumation of these cabins within a comparatively short period after their abandonment.
In Valley-liiver Valley, the writer is informed,3 eleven old shafts have been found, varying in depth from ninety to one hundred feet. In 1854, one of them was cleaned, out, and at the depth of ninety f ;et the workmen found a windlass of post-oak, well hewn,
1 White's "Historical Collections of Georgia," p. 487. Stcplieiison's " Geol ogy niid Mineralogy of Georgia," p. 208. AthmU, Ga., ISVl.
* MS. letter from Dr. Stephcnson.
EVIDE3TCKS OP HAKI/r MIKING.
49
with an inch augur-hole bored through eacli end. Distinct traces appeared where it had been banded with iron. The crank and gudgeon-holes were still in excellent preservation. Another shaft, for twenty-five feet, passed through gneiss-rock. Its sides were scarred by the marks of the sharp tools used in for cing a passage through, this hard substance. There were no signs of Wasting. Below the water-level the casing-boards and timbers were sound, although dis colored by the sulphurets of copper and iron.
Six miles southeast of this locality are five other shafts similar in age and construction. The trees growing in the mouths and upon the edges of these abandoned pits were not less than two hundred years old.
The presence of iron and the marks of sharp metal lic tools prove that these ancient mining operations cannot be referred to the labors of the Indians. The narratives of the career of De Soto are filled with ac counts furnished by the natives of the pi'eseiice of gold in certain designated localities, and their exaggerated statements continually inflamed the cupidity of the adventurers who accompanied the Adelantado on his wild march from Puerto del Espiritu Santo to the broad prairies beyond the Mississippi. In plate xli. of the " BrevisNai'ratio " De Bry presents an extravagant and evidently imaginary illustration of the manner in which the natives gathered gold in the streams issuing from the Apalatcy Mountains. These gold and silverbearing mountains--if we rightly interpret the con fused map accompanying the work to which we have just alluded--were situated, somewhere in or near the northeastern part of Georgia, There is every reason to believe that De Soto passed through Nacoo-
50
ANTIQUITIES OF THE SOUTHERN ISDIAX8.
ehee galley and thence pursued his wanderings by way of tlie Oostenaula or Etowah Valley to their conflu ence. There stood the ancient village of Chiaha, and there now stands the beautiful town of Rome.
"While lingering among the mountains and valleys of North Carolina and Georgia, earnest an.d repeated inquiries were made by the Spanish adventurer re specting the existence of precious metals in that region. Parties were dispatched by him to examine the conntry and ascertain the precise places where the Indians were said to be engaged ill mining. While it does not appear from any of the narratives that De Soto arid his followers actually -underfoot any mining operations-- other than perhaps a limited examination of the surface of the ground--or- that they had with them tools and mechanical appliances which would have enabled them to have penetrated the bowels of the hills and utilized the ores which they contained, it is quite evident that they recognized this as an auriferous region and were greatly disappointed at their failure to secure a consid erable quantity of the coveted treasure.
The question still recurs, Who sunk these shafts, and, in that early day, expended so much labor in ear nest quest for gold ? Dr. Brinton, in an article pub lished in the Historical J^tagazinef has collected, some authorities which suggest a probable response to the inquiry.
So carried away was Lids de Velasco with the rep resentations made by the returned soldiers of De Soto's Expedition, with regard to the gold, silver, and pearls abounding in the province of" Cosa," that he dispatched his general, Tristan de Luna, to open communication with Cosa by the way of Peiisacola Bay. Three hun-
1 I'h-a Series, vo\ r., p. 137.
^r GEOKGIA.
51
tired Spanish soldiers of this expedition penetrated quite to tlie valley of the Coosa, in Northern Georgia and there passed the summer of 1560. Juan Pardo was subsequently sent by Aviles--the first Governor of Florida--to establish, a fort at the foot of the moun tains northwest of St. Augustine, in the province of the chief Coaba. It would seem, therefore, that the Span iards both knew and endeavored, at this early period, to avail themselves of the gold deposits in Upper Georgia. The German traveller, Johannes Lederer, who visited North Carolina and Virginia in 1069 and 1670, and AYrote an account of his adventures in Latin, asserts that the Spaniards were then -working gold and silver mines in the Appalachian Mountains. He avers that he saw specimens of the ore among the AVestern tribes, and brought samples of it bacfe -with him. ''Had I hud with me," he adds, " half a score of reso lute youths who would have stnclv to me, I would have pushed on to the Spanish mines."
In 1690, -while mating a journey over the " Apalathean Mountains" for inland discovery and trade with the natives, Mr. James Moore was informed by the In dians that the Spaniards were at work upon mines within twenty miles of the place where he then wras. The Indians described to him the bellows and furnaces used by these miners, and offered to conduct him to the spot. A difference between himself and his guides, however, prevented his visiting these mines. 1 Subse quently Mr. Moore volxinteerecl to lead a party to these mines, but the scheme fell through.
These authorities, if they do no more, intimate that in the seventeenth century it was believed that the
1 " Collections of the South Carolina Historical Society," vol. i., }>. 209. Charles ton, ISoY.
02
A3TTIQOTTIES OF THE SOUTHERN ISDIAXS.
Spaniards were at work in this region seeking earnestly for gold, and enable us to account, with at least some degree of probability, for tliose physical traces of an cient mining observed by the early settlers of Upper Georgia--operations of no mean significance, and which cannot reasonably be ascribed either to the Indians or to the followers of De Soto.
Returning from this digression, we wonld state that in the manufacture of articles of stone, bone, shell, wood, skin, and feathers, the ingenuity and skill of the primitive workmen found varied and interesting expression. The stone period is here richly repre sented. We have both chipped and polished stone implements of unusual diversity, beautiful material, and most creditable workmanship. A comparison be tween the tastes and labors of the Southern and North ern Indians in this, as well as in almost every other respect, results most favorably to the former. As our attention will, in subsequent chapters, be specially directed to an examination of these various articles and implements, it is necessary here only- to allude to the existence of spear and arrow points, pipes--plain, bird, and animal-shaped--axes grooved and nngrooved, perforated and ornamental, chipped and ground-- gouges, chisels, awls, knives, scrapers', smoothi^g-stones, mortars, pestles, crushing-stones, net-sinkers, tubes, pendants, gorgets, pins, sling-stones, discoidal stones, nut-stones, images, and numerous other articles. In their manufacture, flint, jasper, quartz, chalcedony, slate, steatite, hornblende, diorite, greenstone, soapstone, graywacke and hematite were principally em ployed. Great pains were often expended in their construction. For their pipes and discoidal stones the Cherokeas were famous. Many of the axes, and ar-
PRIMITIVE CANOE UNEARTHED.
Oo
row and spear heads, are marvels of symmetry and beauty. The attention of the workers in shell was mainly directed to the manufacture of heads, headornaments, gorgets, armlets, wampum, pins and per forated disks. Upon the ornamentation of the gor gets nYUch labor and ingenuity were bestowed. Pearls, obtained from salt-water shells and. the fluviatile and lacustrine unionidce, were perforated 'by means of heated copper spindles, and strung and worn around the neck, arms, wrists, waist and ankles.
Plates of mica were used as looking-glasses, and for the ornamentation of the walls of drinking-cups. In the latter case, circular, square, oval, arid diamondshaped pieces were pressed in the clay while still soft --the edges being slightly embedded. "When the ves sel became hard, their retention was insured.
Boats--some of them large enough to convey forty persons--were made of the trunks of trees. The tree was felled, cut off at the desired length, and. hollowed out "by fire. Through its agency also, its sides were shaped, and both the interior and exterior of the canoe scraped and smoothed by means of shells and handaxes or gouges. Bark canoes were seldom if ever used.' They belong to colder waters.
In 1845, while digging a canal on one of the riceplantations, on the Savannah River, located only a few miles distant from the city of Savannah, at a depth of three feet and a half below the surface of the
1 Smith's "History of Virginia," vol. i., p. 132. Richmond reprint, 1819.
54
ANTIQUITIES OF THE SOUTIIKB2T IKDIAXS.
swamp, the workmen came upon a canoe embedded in tlie soil. It answered to the description of what is familiarly known as a dug-out, and had been fashioned from the trunk of a cypress-tree (see illustration). A.bout eleven feet long and thirty inches wide, its depth was scarcely more than ten inches. Both bow and stern were strengthened, each toy a wooden torace kept in position toy wooden pins passing througli the sides of the canoe and entering the toraces at either end. This boat curved upward at either end, so that the toow and stern rose atoove the middle portion. Lo cated about three feet from the stem was seat nine inches wide, consisting of a rude cypress-plank. For its reception the sides of the canoe had toeen notched three inches toelow the gunwales, and it was further kept in position toy four wooden pins--two on each side--driven through the tooat and entering the seat at either end as in the case of the toow and stern toraces.
The bottom was flat, the sides rouiicling. No ef fort had been made to form a keel. The toow and stern, were both pointed, and ziofc tmlike in their gen eral outlines, the latter toeing more tolunt than the for mer. At the top the sides were rather more than half an inch in thickness--increasing, however, as they descended and curved below the water-line.
AVhen cleaned and dried, this canoe weighed sixty pounds, and could be transported with the greatest facility toy a single individual. The agency of fire had obviously been invoked in the construction of this lit tle boat. "While there were no marks of sharp cuttingtools, the evidence appeared conclusive'that the charred portions of the wood, both within and without, had been carefully removed toy rude incisive implements, probably of shell or stone. The plan of felling the tree
MAKTJFAOTUTIE OF ANCIENT CANOES.
OO
and of hollowing out the log, as perpetuated in one, of Do Bry's illustrations,1 seems to have "been observed in this instance. Regarding the regularity with which the outlines and the relative thicknesses of the sides of this boat had been preserved, one could but admire the care and skill with which that dangerous element, fire, had been made subservient to the uses of the primitive boat-builder. It is entirely probable that the ordinary stone celts, chisels, gouges, scrapers, or simple shells, were the only implements at command for the removal of the charred surface, a.s the cypresstree was by degrees converted into the convenient dugout.
In all likelihood., this scraping was done with a shell. Such is the intimation given in an early ac count of the manufacture of canoes by the Virginia Indians: " Mira est in Virginia cymbas fabricandi ratio; nam cum ferreis instrumentis ant aliis nostris similibus careant, eas tameii parare norunt nostris non minus commodas ad nauiqaudum cjuo lubet per flumina &. ad piscaiiduiu. Prjmum arbore aliqua crassa & alta delecta, pro cymba> quarn parare volunt inagnitudiue, ignern circa ems radices stimrna telInre in ambitu struunt ex arbore nrasco bcne resiccato & ligni assnlis paulatim ignem excitantes, lie flamma altius ascendat & arborig longitudiiiein minuat. Fsene adiista <fe ruinani niinante arbore, iiouum suscitant ignem, quern flagrare sinunt donee arbor sponte cadat. A.dustis deiude arboris fastigio & ramis vt truncus iustam longitndinem retiiicat, tignis transnersis supra furcas positis, iinponunt, ea altitudine vt commode laborare possint, tune cortice conehis quibusdam adempto, integriorem truiici partem
56
ANTIQUITIES OP THE SOUTIIEEJT IKDIASS.
pro cymbse inferiore parto seruant, iu altera parte igiiem secundum tranci longitudiiiem struunt, prgeterquam extremis, quod satis adustum illia videtur, restincto igne cooliis scabunt, & nouo suscitato igne deiiuo adurunt, atque ita deinceps pergunt, subinde urentes & scabontes, donee cymba necessarrum aluevun iiacta sit. Sic Domini spiritus rudibtis homiiiibus suggerit ratioiicm qua res in snum usum neccssarias
eonficere queant." ' This canoe had evidently lain for a very long time
in its present position, and seemed to have settled gradually. There was an accumulation of forty inches of mud and soil above it, and around lay the rotting trunks, arms, and roots of forest-trees which, during the lapse of years, had died and become intermingled with the debris of the swamp. Above the spot were growing cypress-trees as large and. seemingly as old as any in the surrounding forest.
It is difficult to form a satisfactory estimate of the age of this relic. That embedded cypress is, for an almost indefinite period, wellnigh indestructible by ordinary agencies, is capable of proof. "We have but to instance the salt-marshes along the line of the Georgia coast, in not a few of which, at the depth of several feet below the surface, may still be found the clearly-defmed and well-preserved traces of cypressforests, consisting of limbs, trunks, knees, and roots. In former years, at least some of these salt-marshes nnist have been fresh-water swamps ; and, without the violent intervention of some marked convulsion of Mature, of which we have no record, and for which no plausible reason can be assigned, centuries must have
ANTIQUITY OF THE BURIEB CANOE.
Oi
elapsed before a gradual settling of tlie coast could liave occurred to such an extent as to have admitted the influx of tidal waves converting cypress-swamps into extensive, uniform salt-marshes, destroying the original growth, and finally covering the fallen forests with mud to the depth of several feet.
A^e are not aware that a sufficiently-accurate rec ord has teen kept of the annual deposit of mud from the overflowing waters of the Savannah River, to en able us to derive from this source a plausible conject ure as to the age of this canoe. So many uncertain ties enter into calculations of this character, that in most instances all attempts to arrive at definite results fall far short of satisfactory conclusions. All we know is, that this Indian canoe is old--older than the barge 1 which conveyed Oglethorpe up the Savannah, when, he first selected the home of the Yamacraws as a site for the future commercial metropolis of the colony of Georgia--more ancient, probably, than the statelier craft "which carried the fortunes of the discoverer of this "Western Continent.
So far as our information extends, this is the first and only well-authenticated instance of the exhuma tion of an ancient canoe in this country. It is in just such a locality that we might have anticipated with greatest confidence the existence of such a relic. The general employment of bark and skin in the manufac ture of their canoes by Northern Indians precludes all reasonable hope of finding ancient specimens made of
such perishable materials. The use of the dug-out, like the presence of a stone
axe, or a jasper arrow-point, tells a true story of the art-condition of the people by whom it is made. It is the simplest form of water-craft, and evidences the
08
AXTIQUITIES OF THE SOUTHERN INDIANS.
first effort in the way of navigation. Hence, among barbarous tribes, there is no essential diversity either in the shape of these primitive boats or in the meth ods of their construction.
The * Andaman-islanders have single-tree canoes hollowed out with a ^-shaped axe, and in their labors arc assisted by the action of fire. On the northeast ern coast of Australia, the natives * use boats formed from a single trunk, hollowed out by fire. The Clalau Indians excel in the manufacture of dug-outs made from the trunks of cedar-trees. In the days of Columbus the natives * of San Salvador fashioned their canoes from the trunks of single trees, hollowins? them out by fire and polishing them, with primitive adzes of flint or shell. While passing down the Mississippi, Heimepin * noted the existence, among the natives, of '~piror^ies or lieavy wooden canows made of the trunks of trees and hollowed out with fire."
William Bartram * says: " These Indians (of South ern Florida) have large, handsome canoes "which they form out of the trunks of cypress trees (cupressus disiicJia^, some of them commodious enough to accommo date twenty or thirty warriors. In these large canoes they descend the river on trading and hunting expedi tions to the sea-coast, neighboring islands and keys, finite to the point of Florida, arid sometimes cross the Gulph, extending their navigations to the Bahama isl ands, and even to Cuba; a crew of these adventurers had just arrived, having returned from. Cuba but a few
3 " Prehistoric Times." Sir 3. Lqhbock. Second edition. London, 1869, p. 425.
2 Idem, p. 429. 3 Wilson's "Prehistoric Man," second edition, p. 99. London, 1865. 4 "Xcw Discovery," etc., p. 153. London, 1698. 5 " Travels," etc., p. 225. London, 1792.
AXCTENT CANOES.
OU
clays before our arrival with a cargo of spirituous liq uors, coffee, sugar, and tobacco. One of them politely presented me with, a choice piece of tobacco, which he told me he had received from the Governor of Cuba."
Cabega de Vaca * "bears testimony to the presence of wooden canoes in use among the Indians whom he encountered in his wanderings, but does not allude to the manner in which they were made.
In the narratives of the career of Hernando de Soto in the conquest of Florida, as told by the Knight of Elvas, and related by Hernandez de Bieclina, mention is made of canoes of considerable size and ornament, but "we are not informed as to their precise shape or meth od of construction. They were evidently, however, '. fashioned from the trunks of trees. < Kibault states that the Florida Indians made canoes out of single trees, capable of transporting safely fifteen or twenty persons, and that they "were propelled by short paddles--the rowers standing up right in the boat.
Lieutenant Timberlake," speaking of the canoes in use among the Cherokees, writes : " They are generally made of a large pine or poplar from thirty to forty feet long, and about two broad, with flat bottoms and sides, and both ends alike; the Indians hollow them now (1761) "with the tools they get from the Euro peans, but formerly did it by fire." The "buried ca noes in the valley of the Clyde were generally formed out of a single oak-stem, hollowed out by blunt tools --probably stone axes--aided by the action of fire.
60
ANTIQUITIES OIT THE SOUTHERN
A. few were "cut "beautifully smooth, evidently with metallic tools." " Hence," says Sir diaries I/veil, " a gradation, could be traced from a pattern of extreme rudeness to one showing great mechanical ingenuity." 3r*enicaut affirms that the canoes of the Indians of Louisiana were made l>y setting fire to the foot of a cypress-tree, the fire continuing in the interior until it fell to the ground. " They then burned it off at the desired length. "When, the tree was burned suffi ciently for their purpose, they extinguished the fire with moist earth, and scraped it out with large shells, which are very thick. They then wash them with water in such a manner as to give them a fine polish. These canoes are sometimes twenty-five or thirty feet long, but they make of them various lengths according to the uses for which they are intended." *
From Bossu's " Travels " we extract the following account: " Before the French came into Louisiana the Indians constructed their boats in the following man ner. They went to the banks of some rivers, which arc very numerous in this vast region, and which by their rapidity tear np by the roots the trees which stand 011 their banks. They took their dimensions for length and breadth, and accordingly chose such a tree as they wanted; after which they set fire to it, and as the tree burnt on they scraped away the live coals with a Hint or an arrow, and having sufficiently hollowed it out, they set it afloat. They are very well .skilled in constructing1 these little vessels upon their lakes and rivers. They employ them in time of Avar, and likewise load them with the furs and dried flesh which they bring back from their hunts." B
French's new scries.
London, IT'71.
ANCIENT CANOES.
61
Compared with tlie boats figured by De Bi-y,1 or the ein&aum of lioberihausen, or that taken from the peat-moor of Mereurango, or that found in tile nook of Moringen, as represented in Keller's " Lake Dwell ings," the Savannah liiver canoe is more symmetrical and less trough-shaped than them all, and assimilates more nearly to the form of the modern canoe. The addition of the braces in the bow and stern is unusual, and the presence of the seat is by no means custom ary.
The primitive river-craft of any people, no matter how low in the scale of civilization, is interesting, and, when the former occupants of the soil have passed away, leaving behind them relics at best but few and frail, we experience a sense of genuine satisfaction as we are thus furnished with the physical proof of the precise manner in which the Indians of Georgia con structed the light barks in which they committed themselves to the waters of the Savannah. This rude boat from the Savannah swamp, perhaps the very first ancient American canoe which has been un earthed, confirms our conjectures, and substantially verifies the earliest and most reliable representations which have been preserved of the Indian canoe of the Southern waters.
Shawls, coverings, and articles of dress, were made of feathers, of buffalo, deer, and bear skins, and the hides of other animals, and were woven by hand out of certain fibres. Fishing lines and nets were formed, of the inner bark of trees, and convenient mats and. baskets fashioned with split canes, reeds, and rushes.
Some of the feather mantles were beautifully
1 " Admiranda NaiTatio," plates xii., xiiu Francoforti ud ^rcrriurn. Be Bry, anno 1590. LL Brevis Xarratio," plates x.iii., xlii. Franeoforti ad Mcenuui. De Dry, anno 1591.
()2
ANTIQUITIES OF THE SOUTHERN INDIANS.
wrought, and upon the well-dressed skins of animals were depicted various designs.
Mortars and pestles, bows, spoons, and platters, scats or benches, ornamental posts used in dancing, and huge images were fashioned of wood.
Shell-beads answered, as a medium of exchange or currency. Fire was produced by the vehement collision or rubbing together of two sticks.
They prepared their skins by first soaking them in water. The hair was then removed by the aid of a bone or stone scraper. Deer's brains were next dis solved in water, and in this mixtxire the' skins were allowed to remain, until they became thoroughly satu rated. They were then gently dried, and, while drying, were continually worked by hand and. scraped with an oyster-shell or some suitable stone implement to free them from every impurity and render them soft and pliable. In order that they might not become hard, when exposed to rain, they were cured in smoke, and tanned with the bark of trees. Young Indian-corn, beaten to a pulp, answered the same purpose as the deer's brains."
Laboriously-constructed dams and intricate w~ears were employed in the capture of fish.
In PAINTING and EOCK-WKITING, the efforts of the Southern. Indians were confined to the fanciful and pro fuse ornamentation, of their own persons with various colors, in which'red, yellow, and black predominated, and to marks, signs, and figures, depicted on skins and scratched on wood, the shoulder-blade of a buffalo, or on stone. The smooth bark of a standing tree or the face of a rock was used to commemorate some feat of
PAINTING AND ROCK--WRITING.
Od
arms, to indicate the direction and strength of a mili tary expedition, or the solemnisation of a treaty of peace. High up the perpendicular sides of mountaingorges, and at points apparently inaccessible, save to the fowls of the air, are seen, representations of the sun and moon,1 accompanied by rude characters, the signifi cance of "which is frequently unknown to the present observer. The motive which incited to the execution of work so perilous was, doubtless, religious in its character, and directly connected with the worship of the sun and his pale consort of the night.
Coarsely done and barren of interest, this pictog raphy feebly expresses the rudest attempts at imita tion by means of colored chalks and the pointed frag ment of a flint. Ignorant of phonetic symbols and of letters, the ideographic characters which they employed were such as are more or less common to all semibarbarians." This primitive system of intaglios and picture-writing--designed to convey intelligence and record events--was supplemented by the use of wampuni, of which we will speak more at large hereafter. 3
The art of dyeing feathers, fibres, rushes, and splints of cane and wood, as well as the quills of birds and animals, to be employed in the manufacture of garments, coverings, mats, baskets, and belts, was generally un derstood and practised.
The TRADE RELATIONS existing among these primi tive peoples were extensive. The principal articles of barter were copper, flint and stone implements, pipes,
1 Ilaywood's "Natural and Aboriginal History of Tennessee," p. 113. Nash ville, 1823.
2 &e Ewbank's " North American Rock-writing," p. 8. Morrisania, N. T., 1860. 3 Compare " Aneient Monumeiitsof the Mississippi Valley,' 1 ehap. xviii. Wash ington, 1848. "Journal of the Anthropological Institute of New York," vol. i., p. 5f, ft seq. Bradford's " American Antiquities," p. 1S2. New York, 1843.
64
ANTIQUITIES OF THE SOUTHERN"
shell-ornaments, pearls, and skins. Galena, obsidian, mica, and small masses of native gold and silver also formed subjects of mercliandise. Between the coast and the interior a constant interchange of commodities was maintained. The beautiful jasper and flint arrow and spear points, stone pipes, discoidal stones, and various articles manufactured by the dwellers among
the mountains, were readily sold to the coast-tribes, who gave in exchange for them, shells, pearls, and com modities, native to their region, and held in esteem by those at a distance. The primitive merchantmen en gaged in this traffic were held in special repute, were generously treated, and had at nil times safe-conduct through the territories even of those who were at war with each other. From the same stone grave in Xaeoochee Valley were taken, an ornamental copper axe from the shores of Lake Superior, a large cassia from the Gulf of ^Mexico, and stone weapons made of materi als entirely foreign to that locality. The sepulchral mounds and relic beds contain articles brought from a distance, and very frequently the finest specimens are obtained at the farthest remove from the spot whence the material used in their manufacture was procured. In this circumstance we trace the intervention of the merchantman, and his inclination, even at that remote period, to find special favor in the eyes of his customers.
This early commerce among the !North American Indians is a subject full of interest, and Prof. C. Ran, in his recent article,entitled "Die Tnnachverhaltnisse der Eiiigebornen Nordame.rika's," published in the iirst quarterly number of the fifth volume of the " Archiv fur Anthropologie," has bestowed upon its con sideration much care and research.
CHAPTER III.
Tin: customs obtaining among the Creeks about the close of the last century, with respect to MARRIAGE and DIVORCE, are tlius detailed by Colonel Hawking : * Tlie suitor never applies in person, but sends his sister, mother or other female relative, to the female relations of the woman he desires to secure as his wife. Brothers and uncles on the maternal side, and some times the father, are consulted, but this is simply a matter of compliment, as neither their approval nor opposition is of any avail. If the match is regarded -with favor, a gracious answer is returned to the wom an who made the application. The bridegroom there upon sends a blanket and sxich articles of clothing as he possesses to the females of the bride's family. If accepted, the contract of marriage is concluded, and he may enter the house of his future wife as soon as he chooses. Having built himself a cabin, made a crop and gathered it in, hunted and brought home his game and placed every thing in the possession of his
1 "Sketch of the Creek Confederacy." Collections of the Georgia Historical Society, vol. iii., part 1, p. 73. Savannah, 1S48.
I
I
66
ANTIQUITIES OF THE SOUTHEEK INDIANS.
wife, tlie preliminary ceremony ends, and the woman is bound. From the time of Ids first visit to the house of tlie woman until tlie termination of tlie ceremony, she is completely subject to his will in every particu lar. A man never marries a member of his own tribe. Marriage gives the husband no right over the property of the wife; and, in case of separation, she keeps the children, and all property belonging to them.
Divorce occurs as a matter of mutual consent, or at the choice of either party--the man having the right to marry again at will, but the woman, except during the continuation of the marriage ceremony, being bound until the feasts of the jBoosTcetau of that year are over.
As a general rule, adultery on the part of the fe male only is punished. The matter is taken in charge by the family or tribe of the husband. The members assemble, consult, and determine upon a course of ac tion. If the proof be clear, and they conclude to pun ish the offenders, they divide and proceed to appre hend them. One half goes to the woman's house and the remainder to the family house of the adulterer, or they all go together to eaeh place if they have so re solved. If the offenders are apprehended, they are beaten severely with sticks and then cropped. The hair of the woman is carried in triumph to the public square. If only one of the offenders be taken, satis faction is had of the nearest relative of the party who escaped. If both make their escape, and the family or tribe of the husband return home and lay down the sticks, the crime is forgiven. One family only, the " Wind " (Ho-tul-ul-gee), can take up the sticks a sec ond time. Should the offending parties succeed in ab senting themselves until the Boos-ke-tau is over, they are pardoned, because, at that solemn festival, uni-
ADULTERY AND ITS PUNISHMENT.
67
versal forgiveness is proclaimed for all offences save murder.
In a letter dated the 9th of June, 1733, Mr. Oglethorpe, speaking of the Indians in the vicinity of Sa vannah, says, " They abhor adultery, and do not ap prove of a plurality of wives." He further states that, where adultery had been committed, the injured hus band was entitled to his revenge by cutting off the ears of the adulterer; and, if physically unable to inflict this punishment, he had a right to kill him the first time lie could do so with safety. The Rev. Mr. Bolzius * records the fact that, on the 26th of March, 1734, an Indian (probably of the Yamacraw tribe) cut off both the ears and the hair of his wife, because she had been too familiar "with a white man. This he \avers to have been the usual punishment for adultery in vogue among the Indians in Southern Georgia.
.Adultery among the Creeks, during Captain Homans's 2 sojourn among them, -was punished by severe flagellations, and the loss of the hair, nose, and ears of both parties. Sometimes the man's nose was spared.
Of infidelity in the husband no notice seems to have been, taken, except in cases where he had in fringed upon the vested rights of another of the same sex; and then he was liable only to such punishment as the anger or ability of the injured husband might lead him to inflict.
These marriage customs varied with almost every nation and tribe.3 The intervention of a priest to im-
1 "Extract of the Journals of Mr. Commissary Ton Reck," etc., p. 49. London, 1734.
2 " Concise Natural History of East and West Florida," p. 98. New York, 1775. 3 See Du Fratz's " History of Louisiana," vol. ii., p. 197. London, 1763. Bossu'3 "Travels," etc., vol. i., p. 232. London, 1771. Bartram's "Travels," p. 512. London, 1792. Lawson's " History of Carolina," p. 1S5. London, 1714.
68
ANTIQUITIES OF THE SOUTHERN INDIANS.
part any tiling like solemnity to the contubernal rela tionship, thus established, appears never to have been thought of. The tie--such as it was--originated, in the fancy of the male, was at first sanctioned by the female friends of the woman, and in other cases by the cacique or head men of the tribe, and was dissolu ble at the option of either party.
The Cherokees in the olden time are said to have had no laws against adultery. Speaking generally, it may be affirmed that the Southern Indians were nionogamoxis * for the time being. This, however, was simply a matter of choice, and not of compulsion. The Muscogulges formed a marked exception to this rule. With them polygamy obtained with the utmost latitude--the first wife being esteemed the queen or superior, and the others her hand-maids and associ ates.' "While polygamy was allowed among the Creeks, Captain Romans * declares that it was not usually practised. The only ceremonies attendant upon their marriages consisted in making some presents to the parents of the bride, and in feasting at the hut of the wife's father.
Intermarriages of first cousins was not permitted. If an Indian debauched his sister or any very near re lative, his body was burnt and his ashes thrown into
Bric-keU's " Natural History of North Carolina," p. 304. Dublin, 1737. JTaywood's " Natural and Aboriginal History of Tennessee," p. 276. Nashville, 1823. Adair'a " History of the North American Indians," p. 138. London, 1775. Hennepin's " Continuation of the New Discovery," chap. xvii. London, 1698.
1 " Singuli singulas habent nxores," says De Bry. " Regibus autcm binas ant teriias habcre permissum : Sola tamen primum ducta colitur & pro Reghm agiioscitm."
" Brevis Narratio," p. 4. Francoforti ad Sltonuin, anno 1591. Cabeca de Vaca says : " Every mail has an acknowledged wife. The physk-ians arc allowed more freedom; they may have two or three wives, among whom exist the greatest friendship and harmony."
2 Bartram's " Travels," p. 513. London, 1792. 3 " Concise Natural History of East and West Florida," p. SV. London, IT'75.
ABSENCE OF MORALITY.
69
the river. lie was regarded as tmworthy to remain upon the earth. Among the Carolina tribes the hus band had a right to sell his wife. He changed his wife at pleasure, and had. at the same time as many wives as he was able to maintain. 1
Comparatively little virtue existed among the un married women. Their chances of marriage "were not diminished but rather augmented by the feet that they had been great favorites, provided they had avoided conception during their years of general pleasure. The husband never pretended to recognize any restraint as , imposed by the marital relation, but indulged his fan cies as inclination prompted or opportunity offered. The wife, on the contrary, was deterred, by fear of public punishment, from the commission of indiscretions. Al though these marriages were in great measure tempo rary in their character--constituting alliances of fancy and convenience--it was not uncommon for parties to live together until extreme old age in comparative peace and affection. By the side of the aged Mico Tomo-chi-chi, as thin and. "weak, he lies upon his blan ket, hourly expecting the summons of the pale-king, we see the sorrowing form of his old -wife, Scenauki, bending over and fanning him with a bunch of feathers. 3
In all verity could the Indian husband say of his wife, as Petruchio affirmed of Catherine :
" I will bo master of wliat is mine own ; She is my goods, my chattels ; she is my house, My household stuff, my field, my barn, My horse, my ox, my ass, my anything."
1 Laws-son's "History of Carolina," p. 1ST. London, 1V14. 2 WhHefield's "Journal at Savannah," p. 2. London, 1739. "Historical Sketch of Tomo-chi-chi," by Charles O.Jones, Jr., p. lO1!. Albany, 1868.
70
ANTIQUITIES OF THE SOUTHERN INDIAK3.
Doomed to perpetual drudgery and to that subor dinate position to which woman is always consigned where civilization and religion are not, she was little else than a beast of burden, busied with coolving, the manufacture of pottery, mats, "baskets, moccasins and tunics, a tiller of the ground, a nurse for her own children, and at all times a servant to the commands and passions of the stronger sex. ._.- gjj^jjj Tjavren, passing with great ease through the perils of childbirth, nourishing her offspring from her own breasts, and permitting the child to suck until it was well grown, with her own hands attending to every want of the infant, and guarding well its cleanliness as it lay lashed to its board cradle, it came to pass that the Indian mother seldom had a lame or deformed or sickly child. At an early age the boys were exercised in running, in playing ball, and in the use of the bow and arrow. Prizes were offered for -which they con tended ; and, while quite young, they were made fami liar with the secrets of hunting and fishing. 1
Protracted ceremonies involving isolation, fasting, purgation, self-denial, and ablution, were religiously observed under the personal supervision of the Is-tepuc-cau-chau-thluc-co, or great leader, before the Creek youth was admitted to the dignity and privileges of manhood. Before going to war the young men were compelled, by the observance of certain formalities and prescribed duties, to prepare themselves to receive the war-physic--a charm against all ills.2
Of the COSTUME and OKUAMEKTS of the Southern
COSTUME AND OP.NAME^TS.
*T1
Indiana, tlie following early accounts furnish general descriptions :
As De Soto nearcd Coca, the cacique came out to receive liim at tlie distance of two cross-bow sliots from the town, "borne in. a litter oil the shoulders of his principal men, seated on a cushion, and covered with a mantle of marten-skins of the size and shape of a wom an's shawl. On his head he wore a cliadern of plumes, and he was surrounded by many attendants singing and playing upon flutes.
At Qiiizquiz the great cacique Aquisco, accompa nied ~by two hundred canoes filled "with armed men, waited upon him. These warriors were painted with ochre, and wore great bunches of white and colored plumes. Standing erect in the canoes, they held in their hands bows and arrows and also feathered shields with which they sheltered the oarsmen on either side. The barge conveying the cacique, and those containing his attendant chiefs, had awnings at the poop under which they sat. The cacica of Cutifachique, when she came out of her town to cross the river and extend to the Adeltmtado the hospitalities of her province, was borne to the water's edge in a chair. There she en tered her canoe, over the stern of which was spread an awning. A mat lay extended in the bottom, and above this were two cushions upon which she sat. In the boats which escorted her was carried much clothing of the country, consisting of shawls and skins. These shawls were made, some from the bark of trees and others of feathers, white, gray, vermilion and yellow, rich and suitable for winter. The deer-skins of which moccasins, leggings, and coverings were fabricated, were well dressed and ornamented with many-colored de signs. The cacica wore strings of pearls, one of which
72
ANTIQUITIES OF THE SOUTIIEEN INDIANS.
slis tlirew aroxiRd. De Soto's neclc, exchanging with liim " many gracious words of fri ^clslup and cour tesy." '
Cafoe<ja de Vaca mentions the \ *.. "f an Indian cliief clothed in a painted deer-skin, and borne upon the back of another Indian. Multitudes of his people attended him, some walking in advance and playing1 upon reed flutes/
In plate xxxvii. of the " Brevis ".Narratio," we have a spirited illustration, of the litter in which the chosen queen is being conveyed to the king. The mat, the cushioned seat, the canopy, the long fans of feather, the four chair-bearers with the rods resting upon their shoulders, aud forked sticks carried in the hand to serve as supports to the litter when they paused to re fresh themselves upon the journey--the company of musicians marching in front, playing upon reed flutes, the retinue of female attendants carrying baskets of fruits, and the plumed warriors with javelins in their hands bringing up and guarding the rear--are all rep resented with apparent fidelity. 3
' " Narratives of the Career of Hcrnando de Soto," translated by Buckingham S.nith, pp. 62, 75, 103. New York, 1866.
2 "Relation," etc., translated by Buckingham Smith, p. 31. New York, 18Y1. 3 The explanatory text is as follows: " Dueturus uxorem, Res, ex nobiliorum puellarum ccutu, pulccrriinaai maximainquc deligere jubet : deinde duobus validis longuviia sede aptata, qua; rariorls alic'-yus animalis polio tecta oat, & posteriore ejr.s parte ornata ramis superno nutantibus, ut sedentis caput tegant, Reginam
tincnt, singuli Hgneam furcam manu gerentes, ut longuriis subponant quando quicsccndum csfc ; duo alii utrinqne ad Rcginre latora progrediunturrotunda umbracuia elegantissime confccta in oblongia baculis gcstantes ad Reginam a Soils ardoribua ttieiidam : prseant alii tubas ex arboris cortice eonfectas infiaiites aupevne angustas, inf'erne laxiore?, duobusque dumtaxat foraminibus, supero & infcro, prseditfe, quibus appensEB sunt ovales sphgerulac aura, argent SB, xrete ad majorem concentum, I*orie sequuutur puellse oiniiium formosissinia;, ele^aiitcr ornat;ii torquibus & ar. millis ox margaritis, singulfe can id t rum seleetioribus fructibus plenum manu ferentes, tS;: sub umbilicum supraque coxoiidiees cinetre cert arum arborura musco ad. ob^caiiia tegendum. Eas scquuntur prtotoriani."
BLANKETS.----SHAWLS.----ROBES.
73
The use of this primitive palanquin was com manded only by kings, queens, and the most distin guished personages, and seems to have existed chiefly among the Florida tribes.
AVhile passing through what would now be known as Middle Georgia, De Soto observed blankets among the natives. These, says the Knight of Elvas, re sembled shawls. Some of them were made from the inner bark of trees, and others of a grass 1 resembling the nettle, which, when beaten, becomes like flasr.
Women used them for a. covering, wearing one about the body from the waist downward, and another over the shoulder, with the right arm free, after the manner of the Gypsies. The men, on the contrary, wore but one, which they carried over the shoulder in the same way, the loins being covered "with a bragueiro of deer skin, after the style of the woollen breech-cloth once the fashion in Spain. " The skins," continues the relator, " are well dressed--the color being given to them that is wished--and in such perfection that when of vermilion they look like very fine red broadcloth ; and when black--the sort in nse for shoes--they are of the purest. The same hues are given to blankets." '
Cabeea de Vaca 9 describes mantelets of thread with which the women partially covered their persons.
The most elaborate robe is that depicted in plate xxxix. of the a Brevis Narratio," upon the person of the king as he walks abroad attended by his queen. 'This is said to have been made of the skin of the stag, elegantly prepared and elaborately ornamented with various colors. It is confined in a prominent how or
74
ANTIQUITIES OF THE SOUTIIEBN INDIANS.
knot resting on tlie top of tlie rig]-
'"'er, and
thence falling over the left hip, is supported behind
by a train-bearer. The arms and legs are bare. A cap
is upon the king's head ; his ears are ornamented with
inflated fish-bladders; his elbows, wrists, and knees,
are encircled by beads of shells and pearls, while from
the left shoulder depend three strings of beads of like
material reaching down as far as the right hip, cross
ing the breatt and stomach transversely. Aside from
ter necklace, armlets, and anklets of pearls, and her
ear-ornaments, his queen-consort is devoid of every
covering save the female breech-clout, which differed
from that worn by the men in that it encircled the
hips, or depended from one shoulder, passing trans
versely below the navel and across the opposite hip,
thus in each instance covering the person only in its
most secret parts. Made of moss, it was more flowing
and graceful than the naked flap and band used by
the men to conceal their privates.
"With the exception of these breech-clouts, the
Florida Indians, most of the year, appeared in a state
of nudity. The cold of winter necessitated the use of
shawls and blankets, to which reference has already
been made.
The warriors wore no artificial protection' for
their bodies, but contented themselves with the most
fanciful head-ornaments, and with personal decora
tions. So painted and ochred were their bodies, legs,
and arms, with red, black, white, yellow and vermilion
stripes, that, in the eyes of the Gentleman of Elvas,
these primitive men-at-arms appeared to have on
stockings and doublets. Some wore feathers and
TATTOOING AND SKIN-PAINTING.
(U
others liorns oil their heads. Tlieir faces were b]ackened and the eyes encircled with vermilion to heighten tlieir fierce aspect.1
Children were permitted to go about in an entirely nude condition until, at their own suggestion, hav ing attained the age of puberty, they put on the breech-clout. The male breech-clout is thus described by De -Bry: " Obscceiias partes tegunt cerviiia pelle eleganter parata."
The custom of tattooing existed. " Maxima illoruni pars corpus, brachia, femora pingit elegantibus & conciimis iiguris c[iiarum color numquam obliteratur : in ipsa enim cute sutit impressed noto sive puncturao." Captain Ribault's account of the attire of the Florida Indians is as follows: "The most part of them cover their reins and private parts with fair hart's skinns, painted most commonly with sundry colors ; and the fore-part of their bodies and arms be painted with pretty devised works of azure, red and black, so well and so properly, that the best painter in Europe could not amend it. The women have their bodies paintett with a certain herbe like iinto morse, whereof the cedar trees, and all other trees, be always covered. The men for pleasure do always twine themselves therewith after sundry fashions. They be of tawny color, hawk-nosed, and of a pleasant countenance." '
The coast Indians are represented to have used 1 ess covering than the tribes of the interior. The farther south we observe them, during the warm months of the year, the more scanty seems the attire. A com-
1 " Narratives of the Career of De goto," etc., p. 99. New York, 1888. Com pare plate xiv., " Brcvis Narratio." Francoforti ad Moenum, DC Ery, anno 1591.
a u The Whole and True Diseovevye of Terra Florida," etc., written in Fi-eiicbe ty Captaine Ribaulde, etc., and now newly set forth in the English, the xxx of liay, 1063. Prynted at London by Rjnland Hall, for Thomas Hackett.
76
ANTIQUITIES OF THE SOUTHERN INDIANS.
parison of the plates illustrative of tlie " Admiranda Narratio," with those contained in the " Brevis Narratio," confirms this assertion. Even kings and chiefmen among the Florida tribes mingled with their head warriors upon occasions of state and general delibera tion, with nothing about their persons save the wretched breech-clout/ Among the common people even this was often lacking.
Of the vesture of the Virginia Indians Harlot writes: " They are a people clothed with loose mantles made of deere skins, & aprons of the same, rounde about their middles; all els naked." * In the accom panying plates ' we are made acquainted with the dis tinctive shapes of the female tunic, the priest's cloak fashioned of quilted rabbit-skins, the aprons and tu nics worn by chiefs, the long -winter garments of old men, dressed with the hair on and lined inside with furs, the scanty covering of the conjurer, and the small breech-clout of the boat-maker and fisherman. The following interesting account of the clothing of the Virginia Indians is borrowed from that valuable work, " The True Travels, Adventures and Observa tions of Captaiue John Smith : " " " For their apparell they are sometimes covered with the skinnes of wilde beasts, which in ^Vinter are dressed with the hayre, but in Summer without. The better sort vse large mantels of Deare skins, not much differing ill fashion from the Irish mantels. Some imbroidered with white beads, some with Copper, other painted after their manner. But the common sort haue scarce to cover
* Richmond reprint, 1819, vol. i., p. 129.
ATTIEE OF VIEGINTA INDIANS.
7t
their nakednesse but with grasse, the leaues of trees or such like. We haue seene some vse mantels made of Turky featliers, so prettily wrought and woven with threads, that nothing*could be discerned but the feathers. That was exceeding warme and very hand some. But the women are alwayes covered about their middles "with a skin, and very shamefast to be seene bare. They adorne themselues most with copper beads and paintings. Their women, some haue their legs, hands, breasts and face cunningly imbrex! ered with divers workes as beasts, serpents, artificially wrought into their flesh with blacke spots. In each eare com monly they haue 3 great holes, whereat they hang chaines, bracelets, or copper. Some of their men weare in those holes a small greene and yellow coloured snake, neare halfe a yarel in length, which, crawling and lapping her selfe about his necke, oftentimes famil iarly would kisse his lips. Others weare a dead Rat tyed by the taile. Some on their heads weare the wing of a bird or some large feather with a Rattell. Those Battels are somewhat like the chape of a Rapier, but lesse, which they take from the tailc of a snake. Many haue the whole skinne of a ilawkc or some strange foule, stuffed with the wings abroad. Others a broad peece of Copper, and some the hand e>f their enemy dryed. Their heads and shoulders are painted red with the roote focone brayed to powder, mixed with oyle, this they hold in sommer to preserue them from the heate, and in winter from the ceild. Many other formes of paintings they vse, but he is the most gallant that is the most monstrous to behold."
The shoes of the natives were made of buckskin, reinforced at the bottom. They were fastened, on with running strings, the skin being drawn together like
78
AKTIQT7ITIES OF THE SOUTIIEKN" INDIANS.
a purse, on the top of the foot, and tied round the ankle.'
During the summer the Louisiana Indians Avore Tout little clothing--that of the men consisting of a small apron of deer-skin dressed white or dyed black, the latter color being reserved exclusively for the chiefs. The cloaks of the women were made of the bark of the mulberry-tree, or of the feathers of swans, turkeys, and ducks. The bark of young mulberryshoots was first dried in the sun, and then beaten so as to cause all the woody parts to fall off. The remaining threads were then beaten a second time, and bleached by exposure to the dew. "When well whitened, they "were spun or twisted into thread. Garments were woven, in the following manner. Two stakes were planted in the ground about a yard and a half apart. A. cord was then stretched from the one to the other, to which were fastened double threads of bark. By hand other threads were curiously interwoven, so as in the end to form a cloak about a yard square, with wrought borders round the edges.
Young boys and girls went quite naked. At the age of eight or ten years the girls put on a little fringed petticoat made of threads of mulberry-bark. The boys remained uncovered until they attained a similar age.
" Some women," says Du Pratz,' " even in hot weather, have a small cloak wrapt round like a waist coat ; but when, the cold sets in they wear a second, the middle of which passes under the right arm, and the two ends are fastened over the left shoulder, so
DEE3S OF LOUISIANA INDIAKS.
79
that the two arms are at liberty and one of the breasts is covered. They wear nothing on their heads; their hair is suffered to grow to its full length, except in the fore-part, and. it is tied in a cue behind in a kind of net made of mulberry threads. They carefully pick out all the hairs that grow upon any part of their body."
The shoes of the men and. women were fashioned after the same pattern, and were seldom worn except upon a journej'. They were made of deer-skin, the sole and upper leather being of the same piece, and sewed together on the upper part of the foot. The moccasin was cut about three inches longer than the foot, and folded, over the toes. The quarters were about nine inches high, and fastened round the leg like a buskin. liar-rings of shell, and necklaces " com posed of several strings of longish or roundish kernelstones, somewhat resembling porcelaine," formed the customary female ornaments. With the smallest of these " kernel-stones " they decorated their furs, gar ters, and shoes. In early youth, females were tattooed across the uosc and often down the middle of the chin. Some were pricked all over the upper part of the body, not excepting from the operation even their sen sitive breasts.
During the winter the men covered themselves with a shirt made of two dressed deer-skins, and wore breeches of the same material, which protected the legs. In severe weather a buffalo-skin, dressed with the wool on, was kept next the body to increase the warmth.
The young men were very fond of dress, vying with each other in the decorations upon their vest ments, painting themselves profusely with vermilion,
I
i
SO
ANTIQUITIES OE THE SOUTHERN INDIANS.
wearing bracelets of tlie ribs of cleer--softened in boiHng water, then bent into the required shape, and finally polished so that they resembled ivory--fan cying necklaces like the women, carrying fans in their hands, clipping off the hair from the crowns of their heads and substituting a piece of swan's-skiii with the down upon it, fastening the finest "white feathers to the hairs which remained, and suffering a part of their hair to grow long, so that they could "weave it into a cue hanging over the left ear. Such is the portrait which has been preserved of the Louisiana youthful swells, more than a hundred years ago.
AVarriors who had rendered themselves famous by some gallant exploit, caused a tomahawk to be pricked on the. left shoulder. Underneath was indelibly im printed the hieroglyphic sign of the conquered nation. The figure intended to be pricked was first drawn on the skin, which was then punctured to the depth of the teutli of an inch, and powdered charcoal rubbed in. Marks thus caused were never effaced. Ear-rings were "Worn by the men; and, fastened to their belts, might be seen gourds with pebbles in them.
The chief ornament of the king was a crown of feathers surmounting a black bonnet of net-work fast ened to a red diadem, about two inches broad, em broidered with kernel-stones. The feathers were white, about eight inches tall in front and half as high behind. The women fabricated girdles, garters, and collars for carrying burdens. They also embroidered \vith porcupine-feathers.
Of the habit of the North Carolina Indians, Mr. Lawson ' writes: The winter dress of the women is " a hairy Match-coat in the nature of a Plad. . . At other
1 "History of Carolina,'' etc., p. 190. London, 1714.
ATTIRE OH' CAKOUKA INDIANS.
81
times they have only a sort of Flap or Apron contain ing two Yards in Length and better than half a Yard deep. Sometimes it is a Deer-Skin dress'd white, and pointed or slit at the bottom, like Fringe. When this is clean, it becomes them very well. . . .
" All of them, when ripe, have a small String round the. "Waste, to which another is tied and conies between their Legs, where always is a "Wad. of Moss against the Os piibis; but never any Hair is there to be found. Sometimes they wear Indian Shooes or Mog'g'izons, "which are made after the same manner as the Mens are.
" The Hair of their Heads is made into a long Roll like a Horses Tail, and bound round with Honoak or Pore-elan which is a sort of Beads they make of the Conk-Shells. Others that have not this, make a LeatherString serve. The Indian Men have a Match-coat of Hair, Furs, Feathers, or Cloth, as the Women have. Their Hair is roll'd up on each Ear, as the AVomens, only much shorter, and oftentimes a Roll on the Crown of the Head or Temples, which is just as they fancy; there being no Strictness in their Dress. Betwixt their Legs comes a Piece of Cloth,1 that is tuck'd in by a Belt both before and behind. This is to hide their N"akelness. . . . They wear Shooes of Bucks, and some times Bears Skin, which they tan in an Hour or two, with the Bark of Trees boil'd, "wherein they put the Leather whilst hot, and let it remain a little while, whereby it becomes so qualify'cl as to eiidrrre "Water and Dirt without growing hard. These have no Heels, and are made as fit for the Feet, as a Glove is for the Hand, and are very easie to travel in when one is a lit tle us'd to them. . . . Their Feather Match-coats are very pretty, especially some of them which are made ex-
' Or wad of moss. Lsuvron, p. 203.
82
ANTIQUITIES OF THE SOUTHERN INDIANS.
traordinary charming, containing several pretty Figures wrought in Feathers, making them seem like a fine Flower Silk-Shag; and when new and fresh, they be come a. Bed very well, instead of a Quilt. Some of another sort are made of Hare, Raccoon, Bever or Squir rel-Skins, which are very warm. Others again are made of the green Part of the Skin of a Mallard's Head, w]rich they sew perfectly well together, their Thread being either the Sinews of a Deer divided very small, or Silk-Grass. AVhen these are finish'd they look very finely, though they must needs be very troublesome to make. . . . Their Dress in Peace and War is quite dif ferent. Besides, when they go to "War, their Hair is comb'd out by the "Women, and done over very much with Bears Grease and red Root; with Feathers, ^Vings, Rings, Copper and Pealc or Wampum, in. their Ears. Moreover, they buy Vermillion of the Indian Traders, wherewith they paint their Faces all over red, and commonly make a Circle of Black aboxit one Eye, and another Circle of "White about the other, whilst others beclawb their Faces with Tobacco-Pipe Clay, Lamp-black, black Lead and divers other Colours which they make with the several sorts of Minerals and Earths that they get in diiferent Parts of the Country where they hunt and travel. When these Creatures are thus painted, they make the most frightful Figures that can be imitated by Men, and seem more like Devils than Humane Creatures. You may be sure that they are about some Mischief, when you see them thus painted ; for in all the Hostilities which have ever been acted against the English at any time, in several of the Plan tations of America, the Savages always appear'd in this Disguize, whereby they might never after be discovered or known by any of the Christians that should happen
OK^AM-M^-TTS OF CAROLINA IKDIAKS.
83
to see them after they had made their Escape ; for it is impossible ever to know an Indian under these Colours, although he has been at your House a thousand times, and you know him at other times as well as you do any Person living. A.S for their AVomen, they never use any Paint on their Faces. . . .
" Some of the Indians wear great Bobs in their Ears, and sometimes in the Holes thereof they put Eagles and other Birds Feathers for a Trophy. When they kill any Fowl, they commonly pluck off the downy Feathers and stick them all over their Heads. Some (both Men and AVoinen) "wear great Necklaces of their Money, made of Shells. . . . They oftentimes make of this Shell a sort of Gorge, which they wear about their Neck in a String; so it hangs on their Col lar, whereon sometimes is engraven a Cross, or some odd sort of Figure which conies next in their Fancy." '
De P>rahm 2 asserts that the South Carolina tribes, about the middle of the last century, had, among themselves, no distinction of dress. They painted their faces red in token of friendship, and black, in expression of warlike intentions. In common with their more northern and southern neighbors they or namented their hair, ears, and necks with feathers, bobs and beads, wore mantles and breech-cloths, and used leather inacksins. " Their cloathing," says Thom as Ash, consists of the " Skins of the Bear and Deer, the Skin drest after their Country Fashion, sometimes with black and red Clieguers coloured." 3
Captain Bernard Romans observed cloth made out
1 Compare " Xatural History of North Carolina," etc., by the wonderful plagia rist, John Brickell, M. D., p. 812, ft aeq. Dublin, 1737.
a " Documents connected with the History of South Carolina," edited by riowden Charles Jennett AVeston, p. 220. London, 1856.
" " Carolina," etc., by T. A.--, Gent., p. 35. London, 1682.
84
ANTIQUITIES OF THE SOUTHERN INDIANS.
of the tart of a species of Morus, the root of the tree being used to dye it yellow. " Buffaloe's "wool," he adds, " furnishes a material for a useful manufac ture. They likewise make "blankets and other cover ings out of the feathers of the breasts of wild turkies by a process similar to that of our wig-makers when they knit hair together for the purpose of making wigs." '
Daring the spring of 1811, embedded in the flooring of a copperas cave, in "Warren County, West Tennessee, two human bodies--the one male and the other female --were found. They were evidently Indians, and had been interred in curiously-wrought baskets made of cane, with coverings of the same material fitting over their tops. " The flesh of these persons," says Mr. Haywood,* " was entire and undecayed, of a brown, dry ish colour produced by time, the flesh having adhered closely to the bones and sinews. Around the female, next her body, was placed a well-dressed deer skiu. Next to this was placed a nig, very curiously wrought, of the bark of a tree and feathers. The bark seemed to have been formed of small strands, well twisted. Around each of these strands feathers were rolled, and the whole woven into a cloth of firm texture after the manner of our common coarse fabrics. This rug was about three feet wide and between six and seven feet in length. The whole of the ligaments thus framed of bark -were completely covered by the feath ers, forming a body of about one-eighth of an inch in thickness, the feathers extending about one-C|uarter of an inch in length from the strand to which they were confined. The appearance was highly diversified by
HUM A If BODIES IN A CAVE I^f TENUESSEE.
85
green, blue, yellow and black, presenting different shades of color when reflected upon by the light in dif ferent positions. The next covering was an undressed deer skin, around which was rolled, in good order, a plain shroud mamifactured after the same order as the one ornamented with feathers. This article re sembled very much in its texture the bags generally used for the purpose of holding coffee exported from the Havaima to the United States. The female had in her hand a fan formed of the tail feathers of a turkey. The points of these feathers were curiously bound by a buckskin string, well dressed, and were thus closely bound for about one inch from the points. About three inches from the point they were again bound by another deer skin string, in such a manner that the fan might be closed and expanded at pleasure. Between the feathers and this last binding by the string, were placed, around each feather, hairs which seem to have been taken from the tail of a deer. This hair was dyed of a deep scarlet red, and was one-third, at least, longer than the hairs of deer's tail in this climate gen erally are.
The male was interred sitting in a basket, after' the same manner as the former, with this exception, that he had no feathered rug, neither had he a fan in his hand. The hair, which still remained on their heads, was entire. That of the female was of a yel low cast, and of a very fine texture, . . . The fe male was, when she deceased, of about the age of fourteen. The male was somewhat younger. The cave in which they were found abounded in nitre, copperas, alum and salts. The whole of this covering-, with the baskets, was perfectly sound, without any marks of decay."
80
ANTIQUITIES OF THE SOUTIIEKN INDIANS.
Thus have the conserving properties of the dust of this cave guarded from disintegration not only the forms, hut even the clothing of these primitive peo ples, offering them almost unchanged for the inspec tion of a later and not incurious age, placing in our hands the fabrics they wove, the skins they dressed, the colors and fans in which they delighted, and affording physical confirmation of the fidelity of at least some of the accounts furnished by the early "ob servers.
Referring to the tribes then occupying the territory granted to the colony of Georgia, Mr. Oglethorpe, shortly after the settlement of Savannah, declares that " they, as the ancient Grermans did, anoint with oil and expose themselves to the sun, which occasion's their skins to be brown of color. The men paint them selves of various colors, red, blue, yellow and black. The men wear generally a girdle with a piece of cloth drawn through their legs and turned over the girdle both before and behind, so as to hide their nakedness. The women wear a kind of petticoat to the knees. Both men and women in the winter wear mantles something less than two yards square, which they "wrap round their bodies as the Romans did their toga, generally keeping their arms bare ; they are sometimes of woollen bought of the English, sometimes of furs which they dress themselves. They wear a kind of pumps \vhich they call moccasins, made of deer skin, which they dress for that purpose." *
" Formerly," says Adair,2 " the Indians made very handsome carpets. They have a wild hemp that grows about six feet high in open, rich, level lands,
1 Salmon's " Modern Histor " Hiatoi-v of the American
1
SPINNING AND WEAVING.
8I
and which usually ripens in July; it is plenty on our frontier settlements. When it is fit for use, they pull, steep, peel and beat it; and the old women spin it off the distaffs with wooden machines, having some clay on the middle of them to hasten the motion. When the coarse thread is prepared, they put it into a frame about six feet square, and, instead of a shuttle, they thrust through the thread with a long cane, having a large string through the web, which they shift at every second course of the thread. When they have thus finished their arduous labour, they paint each side of the carpet with such figures of various colours as their fruitful imaginations devise; particularly the images of those birds and beasts they are acquainted '.with; and likewis eof themselves, acting in their social and martial stations." lie was informed that the Muscogees, time out of mind, passed the woof with a shuttle, " having a couple of threddles which they move "with the hand so as to enable them to make good dispatch, something after our manner of weav ing." The \vomen were the manufacturers of these fabrics. Buffalo's wool was extensively used for spin ning and weaving. Thev Choctaws made " turkeyfeather blankets with the long feathers of the neck and breast of that large fowl." The inner end of the feather was twisted and made fast in a strong double thread of hemp or coarse twine made of the inner bark of the mulberry-tree. These threads were then worked together after the manner of a fine netting. The long and glittering feathers imparted to the out side of the blanket a pleasing appearance. Such fab rics "were quite warm. This writer also confirms the
88
ANTIQUITIES OF THE SOUTHERN INDIANS.
use of breech-cloths, short petticoats, moccasins, and head-ornaments of feathers. 1
Various clays, and the juices of roots, barks, berries and plants, were employed in painting their persons and dyeing their manufactures. Tassels of the hair of deer, colored reel, were held in special esteem.2
Not only wore the ears slit for the reception of in flated bladders, eagles' claws, feathers and various ornamental pendants, but in some instances the nip ples and under lips were bored so that canes and other matters for personal adornment might be introduced and worn.8 The nose was perforated to admit of the suspension of ornaments from the cartilaginous wall "which separates the nostrils. It would appear that lip-stones (called by the Spanish Tiezote and by the Mexicans teutetl^) were worn, at least to a limited extent.
Without multiplying these references, we are suffi ciently assured of the fact that, in the ornamentation of their skins, in the manufacture of shell, stone, bone, wood, hair, and feather pendants, and in the fabrica tion of skin garments fringed and curiously colored, and of carpets and shawls made of fibre and feathers, a marked similarity existed among the Southern tribes. It is also evident that the manner of wearing these articles of clothing and of personal adornment was, in its general features, common to them all. "We per ceive that these Indians had advanced beyond that rudest stage when the undressed hide--stripped from the body of the slain wild animal and thrown around
i Travels," pp. 499-502. London, 17S2. Honncpiu'a
abeija d 8 Ibid., pp. 75--7'
EUROPEAK FABKICS EAGKKLY SOT7GHT.
89
the shoulders--constituted the only protection against the inclement seasons, and that, in the manufacture of their feather coverings and. in the decoration of their persons and garments, considerable taste was some times exhibited.
At an early period, the natives recognized the superiority of the European commodities, and eagerly exchanged their coarse fabrics for the strouds, blan kets and trinkets exhibited by the white traders.
f.
I
CHAPTER IV.
Music and Musical Instruments.--Dnncing.--Hames.--Gambling.--Festivals.-- Divisions of the Year.--Counting.--Funeral Customs.
THE Southern Indians were much addicted to OAIIKS, DIVERSIONS, FESTIVALS, and DANCING. It has been quaintly remarked that man is the only animal that laughs, and we find in all ages, and among all peoples, how limited soever their resources, or narrow their avenues to' pleasure, special attention has ever been paid to the subject of pastimes arid amusements. During periods when the physical development and active training of the human body were eminently ne cessary for individual protection, subsistence, and a toleration of the dangers and privations incident to the precarious and exposed mode of life, the games in vogue were decidedly muscular in their character, and were conducted in the open air. On occasions of feasting and dancing, the music, both instrumental and vocal, was of that simple, primitive kind, adapted to mark the time required for the saltatory movements in which the performers indulged. Measured sounds there were, but melody and harmony were wanting. The cane flute, the drum and the rattle, constituted the principal musical instruments in vogue among the Southern
MUSIC A"ND MUSICAL IHSTBUMENTS.
91
tribes. The form of tlie decorated reed-flute or recorder ' has been preserved for our information, and we are also familiar with the shape of the hand-rattle." " For their mnsicke," says Captain John Smith,3 " they vse a tkicke Cane on which they pipe as on a Recorder. For their warres they haue a great deepe platter of wood. They cover the inouth thereof with a skin, at each corner they tie a walnut, which meeting on the backside neere the bottome, with, a small rope they twitch them to gether till it be so taught and stiffe, that they may beat vpon it as vpon a drumme. But their ehiefe instru ments are Rattles made of small gourds or Pumpeon's shels. Of these they haue Base, Tenor, Countertenor, Meane and Treble. These, mingled with their voyces sometimes twenty or thirtie together, make such a ter rible noise as would rather affright then delight any man." Mr. Bartraiu" asserts that the Southern Indians were all fond of music and dancing, the music being both vocal and instrumental. Among the musical instru ments he enumerates the tambour, the rattle-gourd, and a kind of flute made of the joint of a reed, or of a deer's tibia. The last he pronounces a howling instrument, producing, instead of harmony, "a hideous, melancholy discord." With the tambour and rattle, however, ac companied by sweet, low voices, he confesses himself well pleased. These gourd-rattles contained corn, beans, or small pebbles, and were shaken by hand or struck against the ornamental posts which marked the dancing-ring, in exact time with the movements of the performers. Large earthen pots, tightly covered * with
1 " Brevis Narratio," plate xxxvii.
a " Admiranda Narmtio," plate xviii,
3 "True Travels," etc., vol. i., p. 136. Richmond reprint, 1819.
4 " Travels," etc., p. 502. London, 1792. < BriekeM's "Natural History of North Carolina," p. 328. Dublin, 1737.
Beverly says that these earthen drums were half-full of water. " Uistory of Vir
ginia," book iii., p 55. London, 1705
92
ANTIQUITIES OF THE SOUTHERN INDIAN'S.
dressed deer-skins, answered, as drums. The shells of terrapins were also fastened to the ankles or suspended from the waist-belts. These being partially filled with small stones or beans, with every motion of the body, gave utterance to rattling sounds. The leather stock ings of the young dancing-women of the Creeks were " hung full of the hoofs of the roe-deer, in form of bells, in so much as to make them sound exactly like casta-
gnettes." Captain Romans counted four hundred and ninety-
three of these horn-bells attached to one pair of stock ings. Nine women, whose hose "were similarly fur nished, were present at the dance. Allowing the same number of these tinkling ornaments to each, we will perceive, by easy calculation, that one thousand one hundred and ten deer must have been killed"to furnish these women, with their dancing-bells. These musical instruments were supplemented by voices plaintive or vehement, slow or rapid, as best accorded with the character of the dance. Their songs, whether of "war or devotion, harvest or hunting, consisted of but few words and scanty intonations, repeated in the most monotonous way. AVhen. Ave turn to the music and poetry of these peoples, we enter indeed upon a barren field, with, scarcely any thing to provoke inquiry or re ward investigation.
Iii the vicinity of the village was a spot specially prepared for and devoted to the dance. Here a fire was nightly kindled, and all who had a mind to be merry, assembled each evening. 1
In plate xviii. of the " Briefe and True Report of the New-found Land of Virginia, made into English by Thomas Harlot," we have a lively representation of a
"Admiranda Narratio,'' plate xx.
1
DANCING.
03
public dance--the occasion a great and solemn feast, to which the inhabitants of neighboring towns had been invited--the place, a level spot in the midst of a broad plain, circular in shape, about which are planted in the ground posts " earned with heads like to the faces of nonnes couered with thej'r vayles," the centre being occupied by " three of the fayrest Virgins of the companie, which, imbrassinge one another, doe, as yt wear, turne abowt in their dancinge." Around these, and following the line of the posts, fancifully attired, and bearing in their hands the branches of trees and gourdrattles, with which, they keep time by striking them against the posts, are wildly singing and dancing, in. the cool of the evening, the natives assembled for the celebration of this " solemne feaste."
1 Many of these dances were of a purely social char acter, and were participated in every night by way of amusement. Others were designed, by violent exer cise, to prepare the actors " to endure fatigue, and im prove their wind." * Others still -were had in com. memoration of war, of peace and of hunting ; others in the early spring when the seed was sown, others when the harvest was ended ; others--wild, and teirible--in presence of captured victims doomed to tor ture and death ; while others, with slow and solemn movement and carefully-observed ceremonies, "were conducted in honor of some religious festival. There was scarce an occurrence of note, or a convocation of moment, which did not receive commemoration by a dance. Every occasion was provocative of this amuse
ment. Referring to the dancing of the tribes composing
1 Ltiwson'a " Carolina," p. 175. London, 1714.
94
AKTIQITITIE3 OF THE SOUTHERN INDIANS.
the Greek Confederacy, Mr. Bartram * writes: " They have an endless variety of steps, but. the most com mon, and that which I terra the most civil, and indeed the most admired and practised amongst themselves, is a slow, shuffling, alternate step ; both feet move forward, one after the other, iirst the right foot fore most, and next the left, moving one after the other, in opposite circles, i. e., first a circle of young men, and, within, a circle of young women moving to. gether opposite ways, the men with the course of the sun. and the females contrary to it ; the men strike their arm with the open hand, and the girls clap hands and raise their shrill, sweet voices answering an ele vated shout of the men at stated times of termination of the stanzas ; and the girls perform an interlude or chorus separately.
" To accompany their dances they have songs of different classes, as martial, bacchanalian and amorouSj which last, I must confess, are extravagantly libidi nous ;" and they have moral songs which seem to be the most esteemed and practised, and answer the pur pose of religious lectures."
The Choctaws were distinguished above their neighbors for their poetry and music. Between, their towns existed great rivalry in the composition of songs for dances, and each year, upon the solemnization of the Busk, at least one new song was produced.
Captain Smith thus describes a dance made for his entertainment by Pocahontas during the absence of her father: " In a fair, plain Field they made a Fire, before -which he sat down upon a Mat, when suddenly amongst the Woods was heard such a hideous Noise
," etc., p. 503. London, 1V92. Bossu's account of the dance of impudicity. " Travels," vol. i.,
DAXCE OF POCAHONTAS.
95
and shrieking tliat tlie English betook themselves to their Arms, and seized on two or three Old Men by them, supposing Powliatan, with all his Power, was coming to surprize them. But presently Pocahontas came, willing him to kill her, if any hnrt were intended ; and the beholders, which were Men, "Women and Childi-en, satisfied the Captain that there -was no such matter. Then presently they were presented with this Antick : thirty young M^omen came naked out of the Woods, only covered behind and before with a few Green Leaves, their Bodies all painted, some of one color, some of another, but all differing; their Leader had a fair pair of Buck's Horns on her Head and an Otter's Skin at her Grirdle, and another at her Arm, a Quiver of Arrows at her Back, a Bow and Ar\rows in her Hand. The next had in her Hand a Sword, another a Club, another a Potstick; all of 'em being Horned alike. The rest were all set out with their several Devices. These Fiends with most Hel lish Shouts and Cries, rushing from among the Trees, cast themselves in a Ring about the Fire, Singing and Dancing with most excellent ill variety, oft falling into their infernal pa,ssions, and then solemnly betaking themselves again to Sing and Dance; having spent near an hour in this J^Tascarodo, as they enter'd, in like manner they departed."
In plate xxxviii. of the " Brevis !Narratio," we see nineteen of these dancing-girls moving in a circle and singing the praises of the king and queen. Their steps are more graceful and their motions far less violent and irregular than those practised in religious dances, such, for example, as were observed upon the occasion of the sacrifice of the first-born. 1
96
ANTIQUITIES OP THE SUUTHEKN" INDIANS.
The great game upon which the Southern Indians staked both personal reputation and property, was the c7iuny7ce-gayne. It was played by the warriors, and with those discoidal stones, the symmetry and beauty of which have attracted so much attention. So impor tant was this amusement, so general the indulgence in it, and so desperate the betting, that we have deemed it proper to devote a separate chapter to its history and conduct.
In ball-play one village or tribe was often arrayed against the other, and the contest, although generally good-natured, was prosecuted with so much vigor and excitement, that the players sometimes encountered blows and tumbles which entailed severe bruises and broken limbs. This game was esteemed noble and manly; and, in its exercise, involved feats of strength and agility. Youths of both sexes were frequently engaged, and the principal matches were had in the fall of the year. One chief challenges another to the contest. They meet and make up the game, each se lecting from his own tribe an equal number of contest ants. Upon the appointed day the respective parties meet and lay off the ground upon some plain agreed upon, in the vicinity of a town. Huch property is staked upon the issue, and this is deposited in a pile. Each party is then addressed by its chief, who admon ishes fair play and animates the contestants with the hope and glory of beating their antagonists. The chiefs take no active part in the sport, but, occupying a suitable position, act as judges. The players arrange themselves in the centre of the ball-ground, and the game proceeds. From several accounts descriptive of the manner in which the game was played, we select the
BALL--PLAT.
97
following, furnished by Mr. Adair : J " The ball is made of a piece of scraped deer-skin, moistened, and stuffed hard with deer's hair, and strongly sewed with deer's sinews. The ball-sticks are about two feet long, the lower end somewhat resembling the palm of a hand, and which are worked with, deer-skin thongs. Be tween these they catch the ball, and throw it a great distance, "when not prevented by some of the opposite party, who fly to intercept them. The goal is about five hundred yards in length ; at each end of it they fix two long, bending poles into the ground, three yards apart below, but slanting a considerable way outwards. The party that happens to throw the ball over these, counts one ; but if It be thrown underneath, ;.it is cast back and played for as usual. The gamesters are equal in number on each side; and at the begin ning of every course of the ball, they throw it up high in the centre of the ground, and in a direct line be tween the two goals. "When, the crowd of players pre vents the one who catched the ball, from throwing it off with a long direction, he commonly sends it the right course by tin. artful, sharp twirl. They are so exceedingly expert in this manly exercise, that, between the goals, the ball is mostly Hying the different ways, by the force of the playing sticks, without falling to the ground, for they are not- allowed to catch it with, their hands. It is surprising to see how swiftly they fly, when closely chased by a nimble-footed pursuer ; when, they are intercepted by one of the opposite party, his fear of being cut by the ball-sticks, commonly gives them an opportunity of throwing it, perhaps a hundred yards ; but the antagonist sometimes runs up
1 "History of American Indians," p. 400. London, 1775.
98
ANTIQUITIES OP THE SOUTHERN" INDIANS.
behind, and by a sudden stroke daslies clown the ball. It is a very unusual thing to see them act spitefully in any sort of game, not even in this severe and tempting exercise.
" Once, indeed, I saw some break the legs and arms of their opponents, by hurling them down, when on a descent and running at full speed. But I afterward understood, there "was a family dispute of long continu ance between them, that might have raised, their spleen as much as the high bets they had then at stake, which was almost all they were worth. The Choktah are exceedingly addicted to gaming, and frequently, on the slightest and most hazardous occasions, "will lay their all and as much as their credit can procure." The method- of playing this game did not materially differ among the Southern nations. 1
Foot-ball was also a manly and favorite diversion. These games "were followed, by feasting'and dancing in the public square. Trials of skill "were had. with the bow and arrow, the spear and the club. The Natchez wpmen amused themselves with tossing balls by hand, and in playing a game with bits of cane eight or nine inches long. " Three of these they hold loosely in one hand, and knock them to the ground with another ; if two of them fall with the round side undermost, she that played counts one ; but if only one, she counts nothing.""
Lawsoa " mentions several gambling games, as be-
1 Compare Romans' " Concise Natural History of East aiic. West Florida," p. 79. N"ew York, 1775. Haywood's "Natural and Aboriginal History of Ten nessee," p. 285. Nashville, 1823. Bartram's " Travels," cte., p. 506. Lon don, 1792.
" I>u Pratz' "History of Louisiana," vol. ii., p. 23<J. London, 1763. 3 " History of Carolina," p. 176. London, 1714. Compare Hennepin's " Con tinuation of the New Discovery," etc., chap, xxi , p. 82. London, 1698.
GAMDTG, JFEASTS, ETC.
99
ing In vogue among the Carolina Indians, some played with split reeds and others with persimmon-stones.
To suck a desperate extent was gaming carried, that, having lost all their property, the players would not infrequently stake upon the final issue even their personal liberty, and remain willing servants of the victors until redeemed by relatives and friends.
The great feast of the year, among the Creeks, was the Jioos-Jee-tav. It was celebrated in July or August, and partook of the character of a sacred festival, dur ing which universal thanks were offered to the Great Spirit for the incoming harvest. .All fires were then extinguished, and were new lighted from the spark kindled by the high-priest. It was an occasion of gen eral purification and of universal amnesty for all crimes committed during the year, murder exceptecl. 1
Almost every month nad its peculiar feast or festi val. Among the Natchez the year began with our month of March, and was divided into thirteen moons. ^Vith each new moon a feast was celebrated, receiving its name from the principal fruits gathered or animals hunted. Thus, the first moon was called the J^eer moon and was observed with universal joy as the com mencement of the year. This was followed by the festival of stratoberries. The third moon ushered in the small corn, and was impatiently expected because the crop of large corn seldom lasted from one harvest to the other.
The water-melon feast occurred during the fourth moon, answering to our month of June.
100
ANTIQUITIES OF THU SOTJTHEBK HTDIA3TS.
The flftli moon was that of the fishes. At this time grapes were gathered.
The sixth was known as the mulberry moon. The maize or Great-Corn moon succeeded, and was ren dered remarkable by the most noted festival of the year. The T'lirlcey moon answered to our October, while the ninth and tenth moons "were known respec tively as the livffalo and. Hear moons. It was then those animals were hunted.. The eleventh month was called the cold-meal moon; the twelfth, the cJiestnut moon ; and the thirteenth, the walnut moon.1
If we may believe Adair," the annual feast of love was most carefully observed.
There were festivals in honor of war and of peace, feasts of the dead, of marriage, and for curing the sick, and public ceremonies in adoration of the sun and. in solemnization of various religious rites. When not actively engaged, in Minting, or in warlike pursuits, the time of these primitive peoples was largely spent in feasting and dancing. Beneath mild skies, sur rounded by forests yielding many and nutritions fruits, with few wants in the present and little care for the future, their lives were idly given to amuse ments, and the observance of sundry festivals "whose recurrence constituted the epochal events of the year.
When Cabeca de Vaea asserted that the Southern Indians were ignorant of all time, and made no reckon ing either by the month or the year, his statement was not entirely correct. We have already seen that they divided the year into thirteen moons. They also rec ognized four seasons--the return of the sun, summer, the fall of the leaf, and winter. Of the celestial
1 DTI Pratz' " History of Louisiana," vol. ii., p. 1S5. London, 1763. 8 "History of American Indians," p. 113. London, 1773.
DIVISION OF TIME.----FUNF.K AL RITES.
101
luminaries they took little note except of " the daymoon or sun," and of the "night-sun, or moon." Three divisions were assigned to the day--morning, or " the sun's coming-out," mid-day, and " the sun fallen into the water."
Arguing from the periodicity of their public reli gious feasts, Adair advances the idea that they under stood the division of weeks into seven days. The year commenced with the first new moon of the vernal equinox. Knots of various colors and notched sticks were used to mark the lapse of time. The Cherokees counted as high as a hundred u by various numeral names," while the nations of East and West Florida *' rose no higher than the decimal number, adding units after it by a conjunctive copulative." 1 ^\^e conclude these general observations by an allu sion to the FUNJBRAL KITES observed by the Southern Indians. From the multitude of sepulchral shell and earth mounds still extant along the coast, it is evident that in ancient times the islands and headlands "were densely populated. The variant ages of these tumuli, their internal evidence and many physical facts con nected with them, give assurance that this Indian oc cupancy was long continued. Here the small shellmound formed the common grave of the natives-- the larger earth-mounds being generally erected in honor of chief, priest, or some noted person. The common dead were interred in a horizontal position, sometimes singly, but usually in numbers. The corpses or skeletons, with articles of property, "were, in not a few instances, burnt upon the spot prior to the erec tion of the mound-tomb. In the tumuli of chiefs and priests, however, no evidences of cremation appear. In them the corpse was interred in a sitting posture.
102
AKTIQTJITIES OF THE SOUTHERN" INDIAN'S.
A. thick covering of tenacious clay--enveloping the body like a great, rude, inverted jar--or a light-wood post, firmly driven into the earth, against which the skeleton, or dead body was placed, or to which, when seated oil the ground, it was securely lashed with a grape-vine, or cord of some sort, was sometimes em ployed to keep the corpse in propel' position while the earth was gradually accumulated around and above it.
The custom of depositing with the dead articles of personal property, which, it was believed, would prove of service to them both in their journey toward and in the land of spirits, seemingly prevailed from the earliest times. These sepulchral tumuli are located in the vicinity of the ancient villages and fishing-resorts of the natives. The indications are, that the coast was more densely populated than the other portions of the Southern country, excepting, perhaps, the valley-lands of some of the principal streams. It is entirely proba ble that the natives inhabiting the interior resorted, at cei'tain seasons of the year, in considerable numbers, to the islands and headlands of the Atlantic coast and the Gulf of Mexico, for the purpose of fishing and sub sisting upon the various and abundant supplies of food which the salt-water afforded. This the frequency of grave-mounds and relic-beds amply suggests.
As we leave the sea-shore, and until we encounter the rich valleys of more elevated sections, burial-mounds become more infrequent, and those dedicated to the in humation of the general dead contain a larger number of skeletons than mounds of a similar class located on the coast. In them, so far as our observation extends, evidences of cremation are usually wanting.
Through the pine-barren belt sepulchral tumuli are rarely mot with; and such as are found are located in
OtBAVE--MOTJHDS.
103
the vicinity of deep swamps or near tile rivers where luxuriant forests and abundant waters afforded gen erous supplies of game and iish. In the beautiful al luvial valleys of Upper Georgia, Tennessee, the Caro lines, Alabama, Mississippi, and Louisiana, we again encounter the physical traces of a permanent and ex tensive population. Here we are surrounded with monuments attesting the care and labor expended by these primitive peoples in commemoration of the last resting-places of their dead.
These burial-mounds arc conical or elliptical in form, and vary in size from the small tumulus, whose outline can scarcely be traced,-to barrows quite twenty feet high, and a hundred feet or more in diameter at the base. 1 The practice of entombing the dead in artificial tumuli was abandoned by the Southern Indians very shortly after the advent of the European, and there are good reasons for believing that the custom had fallen into disuse prior to that time. The summits and flanks of many large mounds which were never constructed for burial-purposes, contain, only a few feet below the surface, the skeletons of modern Indians. JSTatural elevations and river-bluffs are frequently filled with graves when there is nothing externally to distin guish them as ancient places of sepulture. It would seem from some of the earliest accounts we possess, that in the sixteenth century and among the Florida tribes only kings and liigh-priests were honored with mound-tombs.
From the absence of burial-mounds in many locali ties which we know must have been thickly settled and occupied for many centuries by the red race, we are led to the conclusion that the construction of septil-
104
ANTIQUITIES OF THE SOUTIIEEN INDIANS.
chral tumuli was limited, and that the common dead --undistinguished by sueli laborious sepulture--'were returned to the bosom of mother earth with frail mon uments marking the places of their final repose.
Even where we possess 110 historic knowledge of the preliminary funeral customs, or of the peoples by whom they were observed, it is curious to note the circumstance that contiguous barrows, similar in out ward appearance, -when opened, reveal different modes of interment. As wave after w~ave breaks upon the beach of the great ocean and then is dissipated into the evanescent foam or returns to the main to be seen and heard no more, each leaving, however, upon the strand its owu sea-shells to tell that the tide "was once there, so during the flight of the lapsed centuries have various tribes swept over the same locality, occu pying it in turn, and, when departing, abandoning to those who came after, manifest proofs of their tempo rary dominion, and of the rites observed by them in the inhumation of their dead.
Within the historic period, the Choctaws main tained the custom of erecting mounds over their dead-- the bodies being reserved in bark and cane coffins and deposited in a bone-house until they had acctumilated sufficiently to warrant the ]abor of a general inter ment. In the early narratives we note a singular absence of all personal observation of sepulchral mound-building, and since our acquaiiitaTice with the manners of the Southern Indians the erection of tumuli above the dead "was seldom attempted by them. Instead of concealing the corpses in the womb of the laboriously-constructed earth and shell mounds, they deposited their dead in cane baskets--having first enveloped them in shawls and mats of native
ETJWEBAL BITES OE THE WATCHEZ.
105
manufacture--and laid them away in caves and crev ices in the rocks, hid them in hollow trees, exposed them upon scaffolds, covered them with logs and stones, submerged them in rivers and lakes, and buried them in graves carefully lined with bark and poles. Of the funeral rites observed by the Southern Indians since the European colonization of this region, we -will be advised by the following references.
Among the Natchez the dead "were either inhumed or placed in tombs. These tombs were located within or very near their temples. They rested, upon four forked sticks, fixed fast in the ground, and were raised some three feet above tlio earth. About eight feet long, and a foot and a half wide, they were prepared, for the reception of a single corpse. After the body was placed upon it, a basket-work of twigs was woven around and. covered with mud, an opening being left at the head through which food was presented to the deceased. When the flesh had all rotted away, the bones -were taken out, placed in a box made of canes and then deposited in the temple. The common dead were mourned and lamented for a period of three days. Those who fell in battle were honored with a more protracted and grievous lamentation.
The demise of a Sun was followed by putting to death large numbers of his subjects, both male and female, that lie might not appear imattendecl in the spirit-world.
In 1725 the Stung Serpent, who was the brother of the Great Sun, died. M. Le Page du Pratz was present on the occasion, and furnishes the following description of -what then occurred: " We entered the hut of the deceased, and found him on his bed of state, dressed in his finest cloaths, his face painted with ver-
106
ANTIQUITIES OP THE SOUTHERN INDIANS.
milion, shod as if for a journey, with his feathei'-crown on his head. To his Tbed were fastened his arms, which consisted of a double-barreled gun, a pistol, a bow, a quiver full of arrows, and a tomahawk, liomid his bed were placed all the calumets of peace he had received during his life, and on a pole, planted in the ground near it, hung a chain of forty-six rings of cane, painted red, to express the number of enemies he had slain. All his clomesticks were round him, and they presented victuals to him at the usual hours, as if he were alive. The company in his hut were composed of his favourite wife, of a second wife, which he kept in another village and visited when his favourite was with child, of his chancellor, his physician, his chief domestic, his pipe-bearer, and some old women, who were all to be strangled at his interment. . . . Soon after, the natives begun the dance of death, and pre pared for the funeral of the /Stimff /Serpent. Orders were given to put none to death on that occasion, but those who were in the hut of the deceased. A child, however, had been already strangled by its father and mother, which ransomed their lives upon the death of the Great /Sum, and raised them from the rank of Stink ards to that of J^Tobles. Those who were appointed to die were conducted twice a day, and placed in two rows before the temple, where they acted over the scene of their death, each accompanied by eight of tlieir own relations who were to be their executioners, and by that office exempted themselves from dying upon the death of any of the suns, and likewise raised them selves to the dignity of men of rank. . . . On the clay of the interment, the wife of the deceased made a very moving speech to the French who were present, rec ommending her children--to whom she also addressed herself--to their friendship, and advising a perpetual
EUNEKAL OP THE STUNG SEKPENT.
107
union between the two nations. Soon after, tlie master of the ceremonies appeared in a red-feathered crown, which half encircled his head, having a red staff in his hand in the form of a cross, at the end of which hung a garland of "black feathers. All the upper part of his body was painted red, excepting his arms, and from his girdle to his knees hung a fringe of feathers, the rows of which were alternately white and red. When he came before the hut of the deceased, he sa luted him -with a great 7too, and then began the ciy of death, iii which he was followed by the whole people. Immediately after, the St'img Serpent was brought out on his bed of state, and was placed oil a litter, which six of the guardians of the temple bore on their shoul ders. The procession then began, the master of the ceremonies walking first, and after him the oldest war rior, holding in one hand the pole with the rings of canes, and in the other the pipe of war--a mark of the dignity of the deceased. ISText followed the corpse, af ter which came those who were to die at the inter ment. The whole procession went three times round the hut of the deceased, and then those who carried the corpse proceeded in a circular kind of march, ev ery turn intersecting the former, until they came to the temple. At every turn, the dead child was thrown by its parents before the bearers of the corpse, that they might walk over it; and when the corpse was placed in the temple the victims were immediately strangled. The Stung Serpent and his two wives were buried in the same grave within the temple; the other victims were intered in different parts, and after the ceremony they burnt, according to custom, the hut of the deceased." '
1 Dii Tratz' " History of Louisiana," vol. ii., p. 216. Londoii, 1763.
108
ANTIQUITIES OF THE SOUTHERN INDIANS.
The Virginia Icings, after death, -were disposed of in the following manner: The body was slit in the back, and through the opening thus made the flesh was removed--the. sinews being left so as to preserve the attachments of the various joints. The bones were then dried, the skin being prevented from shrinking by an application of oil or grease. Subsequently they were carefully disposed in proper order in the skin, the vacuities caused by the removal of the flesh being nicely filled with fine white sand, so as to restore the body to its natural size and appearance. Thus pre pared, the corpse "was laid upon a shelf, raised above the floor, in the building erected for the preservation of the corpses of their kings and rulers. This shelf was overspread with mats. The flesh removed during this rude process of embalming, having been exposed upon hurdles to the sun and thoroughly dried, was sewed up in a basket and set at the feet of the body. In this house of the dead -was set up a Quioccos or idol, as a guard or sacred watcher over the remains. A, priest remained in constant attendance night and day, whose office it "was to keep every thing in order. 1
The common people were buried in the earth in ordinary graves.
Among the Carolina tribes, the burial of the dead was accompanied with special ceremonies--the expense and. formality attendant tipon the funeral, according* with the rank of the deceased. The corpse was first placed in a cane hurdle and deposited in an out-house, made for the purpose, where it was suffered to remain for a day and a night, guarded and mourned over by
FUNKKAL CUSTOMS OF CABOLINA INDIANS.
lOi)
the nearest relatives, with dislievelled hair. Tliose who are to officiate at the funeral, go into the town, and, from the backs of the first young men. they meet, strip such blankets and match-coats as they deem suit able for their purpose, lu these the dead body is wrapped, and then covered with two or three mats made of rushes or cane. The coffin is made of woven reeds, or hollow canes tied fast at both ends. AVheii everything is prepared for the interment, the corpse is carried from the honse in which it has been lying, into the orchard of peach-trees, and is there deposited in another hurdle. Seated upon mats, are there congre gated the family and tribe of the deceased, and invited guests. The medicine-man or conjurer, having enjoined silence, then pronounces a funeral oration, during which he recounts the exploits of the deceased, his valor, skill, love of country, property, and influence, alludes to the void caused by his death, and counsels those who remain to supply his place by following in his footsteps, pictures the happiness he will encounter in the world of spirits to which he has gone, and con cludes his address by an allusion to the prominent traditions of his tribe. He is followed by other speak ers. "At last," says Mr. Lawson," "the Corpse is brought away from that Hurdle to the Grave by four y,7 ouno g Men,' attended byJ the Relations,' the King&,' old Men, and all the ^Nation. When they come to the Sepulcre, which is about six Foot deep and eight Foot long, having at each end (that is, at the Head and Foot), a Light-Wood or Pitch-Piue Fork driven close clown the sides of the Grave, firmly into the Ground ; (these two Forks are to contain a Kidge-Pole, as you shall understand presently) before they lay the Corps
1 " History of Carolina," etc., p. 181. London, 1114.
110
ANTIQUITIES OF THE SOtTTHEKW INDIANS.
into the Grave, they cover the Tbottom two or three times over with the Bark of Trees, then they let down the Corps (with two Belts that the Indians carry their Burdens withal) very leisurely, upon the said Barks; then they lay over a Pole of the same "Wood in the two Forks, and having a great many Pieces of PitchPine Logs, about two Foot and a half long, they stick them in the sides of the Grave down each End, and near the Top thereof, where the other Ends lie on the Ridge-Pole, so that they are declining like the Roof of a House. These being very thick-plac'd they cover them [many times double] with Bark; then they throw the Earth thereon, that came out of the Grave, and beat it clown very firm ; by this Means the dead Body lies ill a Vault, nothing touching him. . . .
" Now, when the Flesh is rotted and moulcler'd from the Bone, they take up the Carcass and clean the Bones, and joint them together; afterwards, they dress them up in pure white dress'cl Deer-skins, and lay them amongst their Grandees and Kings in the Quiogozon, which is. their Royal Tomb or Burial-Place of their Kings and ^S^ar-Captains. This is a very large magnificent Cabin [according to their Building] which is rais'd at the Publick Charge of the Nation, and mam.tain'd in a great deal of Form and Neatness. About seven foot high, is a Floor or Loft made, on which lie all their Princes and Great Men that have died for several hundred Years, all attir'd in the Dress I before told you of. No Person is to have his Bones lie here, and to be thus dress'd, unless he gives a round Sum of their Money to the Rulers for Admittance. If they remove never so far, to live in a Foreign Country, they never fail to take all these dead Bones along with them, though the Tediousness of their short daily
rrNEBAL CUSTOMS OF CAROLINA INDIANS.
Ill
Marches keeps them never so long on tlieir Journey. They reverence and adore this Quioyozon with all tlie Veneration and Hespect that is possible for such a People to discharge, and had rather lose all, than hare any Violence or Injury offer'd thereto. These Sav ages differ some small matter in tlieir Burials; some burying right trpwarcls and otherwise. . . . Yet they all agree in tlieir Mourning, which is to appear every ?fight, at the Sepulcre, and howl and weep in a very dismal manner, having their faces dawb'd over with Light-wood Soot []which is the same as Lamp-blackJ and Bear's Oil. This renders them, as black as it is possible to make themselves, so that theirs very much resemble the Faces of Executed Men boil'd in Tar. If the dead Person was a Grandee, to carry on the Funeral Ceremonies they hire People to cry and la ment over the dead Man. Of this sort there are sev eral that practise it for a Livelihood, and are very expert at Shedding abundance of Tears, and howling like Wolves, and so discharging their Office with abundance of Hypocrisy and Art. The women are never accompanied with these Ceremonies after Death ; and to what AVorld they allot that Sex, I never un derstood, tinless, to wait on their dead Husbands; but they have more "Wit than some of the Eastern Na tions, who sacrifice themselves to accompany their Hus bands into the next AVorld. It is the dead Man's Relations by Blood, as his Uncles, Brothers, Sisters, Cousins, Sons, and Daughters, that mourn in good earnest, the AVives thinking their Duty is discharg'd, and that they are become free when their Husband is dead; so, as fast as they can, look out for another to supply his Place."
The ceremonies attendant upon the sepulture of
112
ANTIQUITIES Of THE SorTIIEEN IXDIAXS.
the Choctaw dead are thus described by Captain Ber nard Romans: * " As soon as the deceased is departed a stage is erected and the corpse laid on it and cov ered with a bear-skin ; if lie be a man of note it is decorated and the poles painted red with vermillioii and bear's oil ; if a child, it is put upon stakes set across: at this stage the relations come and weep, ask ing many questions of the corpse, such as, AVhy he left them 2 Did not his wife servo him well ? "Was lie not contented with his children ? Had he not corn enough ? Did not his land produce sufficient of every thing? "Was he afraid of his enemies 2 etc., and this accom panied by loml howlings; the women, will be there constantly, and. sometimes, with the corrxiptecl air and heat of the sun, faint so as to oblige the by-standers to carry them home; the men will also come and mourn in the same manner, but in the night or at other unseasonable times when, they are least likely to be discovered.
" The stage is fenced round with poles, it remains thus a certain time, but not a fixed space, this is some times extended to three or four months, but seldom more than half that time. A. certain set of venerable old Gentlemen who wear very long nails as a distin guishing badge on the thumb, fore and middle finger of each hand, constantly travel through the nation [when i was there i was told there were but five of this respectable order] th at one of them may acquaint those concerned of the expiration of this period, which is ac cording to their own fancy; the day Toeing come the friends and relations assemble near the stage, a fire is made, and the respectable operator, after the body is
3 "A Concise Natural History of East and West Florida," etc., pp. S3, 90. New York, Itto.
FUHERAL CEREMONIES O3T THE CHOCTAWS.
118
taken down, with his nails tears the remaining flesh off the bones and throws it with the entrails into the foe, where it is consumed ; then he scrapes the "bones and burns the scrapings likewise; the head, being painted red with vermillion, is, with, the rest of the bones put into a neatly made chest (which, for a chief, is also made red), and deposited in the loft of a. hut built for that purpose, and called bone-house; each town has one of these ; after remaining hero one year or there abouts, if he be a man of any note, they take the chest down, and in an assembly of relations and friends they weep once more over him, refresh the colour of the head, paint the box red, and then deposit him to last ing oblivion.
" An enemy, and one who commits suicide, is buried under the earth as one to be directly forgotten and un worthy the above ceremonial obsequies and mourning."
Mr. Bartram's account is substantially the same, save that he intimates there is a general inhumation so soon as the bone-house becomes full of coffins. Then the respective coffins are borne by the nearest relatives of the deceased to the place of interment, where they are all piled ono upon another iu the form of a pyramid, and the conical hill of earth heaped above. The funer al ceremonies are concluded with the solemnization of a festival called the feast of t7ie dead. 1
The Muscogulges buried their dead in the earth--a deep pit, about four feet square, being dug under the cabin and couch occupied by the deceased. This grave was carefully lined with cypress-bark, and in it the corpse placed in a sitting posture. Such articles of
1 See Bill-tram's "Travels," ete., .p. 514. London, 1792. Compare Adair's " History of the American Indians," pp. 183, 184. London, 1775.
114
ANTIQUITIES OP THE SOUTHERN INDIANS.
property as he valued most, were deposited with liim.1 Among the Alibamons--who also buried their dead in a sitting posture--to the suicide was denied the rite of sepulture. lie was regarded as a coward, and his body was thrown into a river.2
The funeral customs of the Chicasaws * did not dif fer materially from those of the Muscogulges. They interred the dead as soon as the breath left the body, and beneath the couch on which the deceased expired.
Lieutenant Timber-lake * intimates that the Cherokees, living upon the banks of the Tennessee, seldom buried their dead, but threw them into the river. Mr. Adair's observations were entirely different. lie as serts ' that when any member of this nation died away from home--if his companions were not closely pur sued--the corpse was placed on a scaffold, covered with notched logs, to protect it from wild beasts and birds. When they imagined that the flesh had been con sumed and the bones become dry, they returned to the spot, enveloped the skeleton in white deer-skins, brought it home, and, having mourned over it, buried it with the usual solemnities. Piles of stones were heaped up to commemorate the spots where fell their distinguished -warriors, and to these rude monuments each passer-by added a stone in token of his apprecia tion of the valor and brave deeds of the deceased.
When a Cherokee died at home, his corpse was at once washed and anointed, brought out of his lodge and placed in a sitting posture on the skins of "wild
1 Bartram's "Travels," etc., p. 513. London, 1792. Romans' ;" Florida," p. 98. Sew York, 1116.
2 Bossu's " Travels through Louisiana," vol. i., pp. 257, 258. London, 1771.
FUNEBAL CEBEMONXES OF THE CHEKOKEES. 115
"beasts, supported "by all liis articles of property dis posed around liim, and with liis face turned westward, as though, looking into the door of the winter-house. A eulogium was then pronounced; and, when the period allotted for mourning had elapsed, the body carried three times aronnd the house, in which it was to be interred, those officiating stopping for half a minute at the completion of each Gil-curt. The reli gious man of the family of the deceased, who walked in front, chanted the funeral-song, in the chorus of which the procession joined.
Mr. Adair was present when a chief -was buried. It would appear that he -was interred beneath the floor of a -winter-house. The preliminary funeral rites kavm,g been performed in the manner just indicated, "they laid," says our observer, " the corpse in his tomb in a sitting posture, with his face towards the east, his head anointed with bear's oil and his face painted red, but not streaked with black, because that is a constant emblem of war and death; he was drest in his finest apparel, having his gun and pouch and trusty hiccory bow, with a young panther's skin full of arrows, along side of him, and every other useful thing he had been possessed of, that, when he rises again, they may serve him in that tract of land -which pleased him best be fore he went to take his long sleep. His tomb was firm and clean inside. They covered it with thick logs, so as to bear several tiers of cypress bark, and such a quantity of clay, as would confine the putrid smell, and be on a level with the rest of the floor. They of ten sleep over those tombs, which, -with the loud wail ing of the women at the dusk of the evening, and dawn of the day, on benches close by the tombs, must awake the memory of their relations very often; and
116
ANTIQUITIES OF THE SOUTHERN INDIANS.
if they were killed by all enemy, it helps to irritate and set on such revengeful tempers, to retaliate blood for blood."
Juan Ortiz--sole survivor, among the Florida tribes, of the expedition of Panphilo de Warvaez, and for twelve long years condemned to slavery in the. " Land of Flowers "--was, by his captors, compelled to stand guard at the temple in which the Indian dead reposed. Upon peril of his life he was forced to watch, lest the wild beasts should come by night and steal away the bodies. The story of his good fortune in delivering from the jaws of a predatory wolf the corpse of an Indian boy, is familiar to the readers of the narratives of De Soto's expedition.
The general respect paid by the natives to their dead, the care exhibited in the proper solemnization of their funeral rites, the private and public exhibitions of sorrow, the expressed belief in the existence of a spirit-world, the effort to furnish the deceased with such articles as would prove most serviceable upon the long journey, and in new and pleasant fields, the jealousy with which they watched over and defended the places of sepulture, and the earnestness and hon or with which they perpetuated the memories of the great when they no longer walked among the living, declare that these primitive peoples--how barbarous soever they, in other respects, might have been--held 110" light thoughts from objects of mortality," drew no " provocatives of mirth from anatomies," and showed no jugglers' tricks with skeletons. Their corpses were never knaved out of their graves to have their skulls made into drinking-bowls, or their bones turned into pipes. In nothing was the character of the Southern Indian worthy of greater commendation than in his
VENERATION EOR GRAVES.
117
veneration for the reputation and the tomb of his de ceased leader, in the solicitude with which lie laid his relative and friend to rest "beneath, the shadows of his native forests and within sight of his own village, and in the vigilance with which he insured the undisturbed repose of the dead of family and tribe.
Truthfully might the returning Indian, as he muses over the deserted and mutilated burial-place of his fa thers, exclaim :
"Tins "bank, in which the dead wore laid, "Was sacred when its soil was ours ;
" But now the wheat is green and high, On clods that hid the warrior's "breast,
And scattered in the furrows lie The "weapons of his rest ;
And there, in the loose sand, is thrown Of his large arm the mouldering hone.
" Ah ! little thought the strong and brave Who bore their lifeless chieftain forth,
Or the young wife, that weeping gave Her first-born to the earth,
That the pale race, who waste us now, Among their bones should guide the plough,"
CHAPTER V.
WHAT Sir Thomas Browne * quaintly styles " the restless inquietude for tlie diuturnity of our memo ries," an ambitious desire to wrest from oblivion the names and graves of such as were famed for feats of arms or remarkable for some individual excellence, an appreciation of the fact that ill the tomb of the dead hero lived recollections which, while they dignified the past, also inspired hope in the present and proved a powerful incentive to future action, and that inclina tion (so natural to the human heart in all ages) to render the most affectionate and honorable sepulture to the departed, have united in causing the erection of some of the oldest and most prominent artificial monu ments extant upon the earth's surface. TJrnal inter ments, burnt relics and earth-mounds, inasmuch as they " lie not in fear of -worms," endure when personal and even national memories have perished. In some of them rest the sui'est .and earliest physical proofs of the antiquity of man. .Amid the depths of forests,
1 " Hydriotaphia."
ANTIQUITY OF EARTH--MOUNDS.
119
where every thing like a history or even a tradition of the peoples who once dwelt beneath their shadows, is, to us of the present (lay, emphatically " in the urn," the curiosity of subsequent ages has, in ancient graves and sepulchral tumuli, caught a glimpse of many things appertaining to a forgotten past, learned lessons of the general pyre, the last valediction, the funeral cus toms, the religious rites and the domestic econorny of nameless nations whose former existence could oth erwise have been scarcely more than, conjectured.
In periods the most remote, the earth-mound seems to have suggested itself as the most natural and en during method . of perpetuating the memory and of designating - the last resting-place of the illustrious dead. The mound at Aconithus, erected over Artachies--the superintendent of the canal at Athos--re mains, to this (lay, a memorial of Persian usage, a pub lic recognition of the ability of that engineer, so famous in his generation, and a proof of the fidelity of Herodotus as an historian. Those mighty tumuli which tower along the bants of the Borysthenes are the tombs of Scythian longs. The neighborhood of the Gygaean Lake, near Sardis, in' Asia Minor, is ren dered remarkable by the presence of circular mounds, among which, perhaps, the most recent is that " prince of tumuli," the tomb of Alyattes, King of Lydia, which for nearly twenty-five hundred years has braved the changing seasons.
Allusions to such structures are not infrequent among the ancient poets. Thus Orestes, when ad dressing the manes of the murdered Agamemnon, says :
" If but some Lycian spear 'neatli Ilium's walls Had lowly laid thee,
A migMy name in tlie Atridan halls Thou wouldst liave made tlice.
120
ANTIQUITIES OF THE SOUTHERN INDIANS.
Then liadst thou pitclied thy fortunes like a star To son and daughter shining from afar!
Beyond the wide-waved sea the high-heaped mound Had told forever
Thy feats of battle, and with glory crowned Thy high, endeavor."
The ceremonies attendant upon the burial of Patroclus are thus commemorated in the " Iliad : "
" The Greeks obey. Where yet the embers glow "Wide o'er the pile the sable wine they throw, And deep subsides the asliy heap below. !N"ext tlie white bones Ills sad companions place, "With tears collected, in the golden vase. The sacred relics to the tent they bore ; The urn a vale of linen covered o'er. That done, they bid the sepulchre aspire, And cast the deep foundations round the pyre. High in the midst they heap the szuelling lied Of rising earth, memorial of the dead."
Tydeus and Lycus were buried under earthen barrows, and Alexander the Great caused a tumulus to be heaped above his friend Hephgestion at a cost of twelve hundred talents. So ancient are some of these earth-mounds that they were old and mysterious in the days of Homer. Even in more polished ages, and in seasons of extreme opulence, the memory of the mound-tomb was not forgotten. Its rude earth dome was seen surmounting a circular arrangement of ex quisite porticos, columns, and decorated walls, facing nearly every degree of the circle, and resplendent in. all the carving and polish which the most beautiful mar ble could receive.1
Apart from monuments which we know to have been erected within the historic period, scattered over
1 See Siny th's "Antiquity of Intellectual Man," pp. 102, 103. Edinburgh, 1868.
ANCIENT TUMULI IN" GEORGIA.
121
the plains, peopling tlie valleys, and crowning' the hills of Europe, Asia, Africa, America, and the islands of the ocean, we find ancient tumuli--abundant and si lent witnesses of the early constructive labors of name less tribes and nations.
More than three hundred years ago, artificial tumnli within the present geographical limits of Georgia attracted the notice of the Spanish adventurers and early voyagers. These physical traces of a popula tion apparently older and more patient of labor than that which they found in possession of the soil, while they excited the "wonder and curiosity of the colo nists, do not appear to have enlisted any careful inquiry, or to have received a minute examination. .The most august of them were dismissed with lit tle more than a bare mention of their existence, and, even where descriptions were attempted, tliey were either so meagre in their outlines as to be almost valueless for the purposes of definite information, or so exaggerated as to savor more of romance than of reality.
At a remove from those who could verify their ob servations by personal examination and careful inspec tion--filled with vague conjectures touching manners and matters entirely novel in their character--in a re gion wild, remote, and abounding with strange scenes, unusual features and but partially-comprehended tradi tions--with imaginations often excited to the last de gree--influenced by extravagant rumors--sometimes investing an occurrence, a suggestion, or an object, with an air of importance far beyond its deserts, and again treating with entire neglect or disdainful words things which were really worthy of specific mention and his toric commemoration, the early narrators compel the
123
ANTIQUITIES OF THE SOUTHERN IXDIANS.
candid reader to receive tlielr relations cum grano scilis.
Since the date of tlieir observations, and even of Mr. Bartram's visit, the winds and rains of many sea sons liave sadly changed tlie appearance of these earthmouuds. Worn away by the elements, marred by the ploughshare, and torn asunder by the curious, many of them have been despoiled of their original proportions. The brandies of the forest-trees wdiich once overshad owed, them are, in not a few instances, no longer out stretched for their preservation, and some have been wholly crushed out of existence by the ti'ead of a statelier civilization.
Making, however, due allowance for such changes, after a somewhat extended and careful survey of these monuments, we cannot resist the impression that the early descriptions are frequently not only over-wrought, but imnatural. "What would now be regarded as an ordinary conical mound has, on more than one occasion, been represented as possessing physical peculiarities of ail unusual and. remarkable character.
Grarcilasso mentions the existence of large artificial tumuli -with precipitous sides, flat on the top, and located in rich valleys', neat the banks of beautiful streams, and says that they were erected for the pur pose of sustaining the houses of chiefs and tlieir fami lies. "Wooden stairways made by cutting out inclined planes fifteen or twenty feet "wide, flanked on the sides with posts and with poles laid horizontally across the earthen steps, afforded the means of ascending to their tops. At the foot of these mounds a square was marked out, around which were built the dwellings of the principal men of the tribe. Outside appeared the wigwams of the common people. A_ disposition to
BAETEAM'S ACCOUNT 01? THE GEOEGIA TUMULI. 123
place the residence of the chief in a commanding posi tion--thereby elevating the cacique above his subjects --and a desire to contribute to his personal security are assigned as motives for the expenditure of so much labor.
"Various are the allusions made by that intelligent and interesting traveller, Mr. William Bartram, to the presence of ancient tumuli within the limits of Georgia. Some of his descriptions are evidently exaggerated, but they are the most minute which have been pre served for our information. From them we select the following.
Above the town of ^Vrightsboro and overlooking the low grounds of the north branch of Little Kiver, he saw " very magnificent monuments of the power and industry of the ancient inhabitants of these lands. ... I observed," he writes, " a stupendous conical pyramid, or artificial mount of earth, vast tetragon ter races, and a large sunken area, of a cubical form, en compassed with banks of earth ; and certain traces of a larger Indian town, the work of a powerful nation, whose period of grandeur perhaps long preceded the discovery of this continent." 1
At Silver Bluff, on the Savannah Eiver, the surface of the ground was rendered remarkable by " various monuments and vestiges of the residence of the an cients ; as Indian conical mounts, terraces, areas, etc., as well as remains or traces of fortresses of regular for mation." "
Near Fort James, which was located not far from the confluence of the Broad and Savannah Kivers, the surgeon of the garrison drew the attention of Mr. Bartram to some Indian monuments " worthy of every
1 "Travels," etc., p. SI. London, 1782. ' Ibid., p. 313.
124
ANTIQUITIES OF THE SOUTHERN INDIANS.
traveller's notice. . . . Tliese wonderful labours of the ancients stand in a level plain, very near tne bank of the river, now twenty or thirty yards from it. They consist of conical mounts of earth, and four square ter races, etc. The great mount is in the form of a cone, about forty or fifty feet high, and the circumference of its base two or three hundred yards, entirely composed of the loamy rich earth of the low-grounds ; the top or apex is flat: a spiral path or track leading from the ground up to the top is still visible, where now grows a large, beautiful spreading Red Cedar (Juniperus Americana) ; there appear four niches, excavated out of the sides of this hill, at different heights from, the base, fronting the four cardinal points; these niches or sentry boxes are entered into from the winding path, and seern to have been meant for resting-places or look outs. The circumjacent level grounds are cleared and planted with Indian Corn at present; and I think the proprietor of these lands, who accompanied us to this place, said that the mount itself yielded above one hundred bushels in one season: the laud hereabouts is indeed exceedingly fertile and productive." '
Having suggested that these tumuli were intended to serve as " look-out towers," having commented upon the fact that such public works would have required the united labor and attention of a "whole nation--circum stanced as the Indians then -were--to have constructed one of them almost in an age, and after describing sev eral smaller mounds " round the great one, -with some very large tetragon terraces on each side, near one hundred yards in length," with surfaces elevated four, six, eight, and ten feet above the ground, our author concludes by hazarding the conjecture that these arti-
MOUSES IX THE CHEKOKEE COUNTRY. 125
ficial elevations -were designed as " retreats and ref uges " from tie swelling tide of the river during sea sons of sndden inundations.
The mounds on the east bank of the Oemulgee River, near Macon, did not escape the observation of Mr. Bartram. Even the lonely mounds along the Alatamaha attracted his attention.
The council-house of the Cherokee town of Co we, he tells us, was a large rotunda capable of accomodating several hundred people. It stood " on the top of an ancient artificial mount of earth, of about twenty feet perpendicular," and--the rotunda itself being rather more than thirty feet high--the whole fabric possessed an elevation of about sixty feet. " It is proper to observe," he continues, " that this mount on which the rotunda stands, is of a much ancienter date than the building, and perhaps was raised for another purpose. The Cherokees themselves are as iguorant as we are, by what people or for what purpose these artificial hills were raised ; they have various stories concerning them, the best of which amount to no more than mere conjecture, and leave us entirely in the dark ; bat they have a tradition common with the other na tions of Indians, that they found them in much the same condition as they now appear, when their fore fathers arrived from the West aud possessed them selves of the country, after vanquishing the nations of red men who then inhabited it, who themselves found these mounts when they took possession of the coun try, the former possessors delivering the same story concerning them : perhaps they were designed and ap propriated by the people who constructed them, to some religious purpose as great altars and temples." '
1 "Travels," etc., pp. 365, 366. London, 1Y92.
- 126
ANTIQUITIES OF THE SOUTHERN" INDIANS.
During the progress of this investigation it mil be perceived that mound-building--which, seems to have been falling into disuse among the Southern Indians prior to the dawn of the historic period--was entirely abandoned very shortly after intercourse was estab lished between the Europeans and the red-men. We will observe, moreover, that these ancient tumuli were, by later tribes, subjected to secondary uses, so that in not a few instances the summits and flanks of large temple-mounds originally designed for religious objects --such as the worship of the sun--were, by the Greets and Cherokees, converted into stockade-forts, Ttsed as elevations for council-lodges and the residences of their chiefs, or devoted to the purposes of sepulture. This can scarcely be wondered at when we remember that many of the nomadic tribes who peopled this region wore unstable in their seats, engaged in ever-recurring and annihilating wars, and constantly yielding to the conquest of more powerful neighbors who, expelling them from some coveted hunting-ground or fishingresort, possessed themselves of the desired domain, caring little for the frail memories which clustered about the name and monuments of the vanquished. In an age entirely devoid of letters, it is not surprising that with the lapse of time the victors should have pre served not even a distinct tradition of the conquered. It will be remembered that the North American In dian was generally quite reticent as to his people and their old customs, and frequently denied to the stranger a knowledge of matters which he did not desire either to discuss or to reveal. When we reflect upon the care less and uncertain manner in -which the annals of these peoples were perpetuated, it is not improbable that in the course of centuries all definite accounts of the
ABSEKCE OF M-EGAUTIIIC MONTJMT5MTS.
12 I
builders of these artificial elevations and the history of their construction should have faded from the recol lection even of the descendants of those by whom they were erected.
In one of his addresses to the pupils of the Royal Academy, Sir Joshua Reynolds remarked that -when the ignorant inhabitants of the East were questioned concerning the stately ruins which filled their land-- melancholy monuments of former grandeur and longlost science--their universal response was, "They were built by magicians." Finding a vast gulf between, its own powers and works indicative of skill and great labor, the untaught and inert mind of the savage dis misses the contemplation of their origin and primal uses either with an avowal of utter ignorance on the subject or by referring their creation to the agency of some supernatural influence. It is proper, therefore, to re ceive with caution the traditions delivered by the modern Indians with regard to the erection and history of the more august tumuli which dignify the valleys and tower along the banks of some of the principal rivers in Georgia. "With the exception of stone graves, rock-piles, and walls loosely constructed of stones, laid one upon the other, there is, in this State, a remarkable absence of megalithic monuments, such as dolmens, menhirs, and avenues, -which abound in so many por tions of the Old "World. We search in vain for animalshaped mounds ; and yet Georgia, in almost every sec tion, teems with vestiges of an ancient population now wholly extinct within her borders. Stone tumuli and rudely-constructed rock-walls rear their heads even upon the summit of lofty Yonah. The spurs of the Blue Ridge give frequent evidence of inhumations whose mouldering heaps have for generations defied
128
ANTIQUITIES OF THE SOUTHERN INDIAKS.
the annihilating influences of the tempest. The beau tiful valleys of jSTacoochee, of the Etowah, the Oosteiiaula, the Ohattahoochee, and other streams, are ren dered remarkable by the presence of tumuli of unusual size. Upon the banks of the Savannah, by the -waters of the Ogeechee, and within the swamps of the Alatamaha, are found surprising monuments of ancient industry and devotion. Even throughout the lonely pine-barren region similar remains exist wherever a traant stream or moss-clad swamp infuses new vigor into the forest growth, and affords friendly cover for game. The coast and the low-lying islands are literally studded with tumuli beneath which the unnumbered and nameless dead of centuries repose.
As the presence of these mounds may be regarded as indicating the particular localities most thickly peo pled, by the aborigines in years long since reckoned with an unrecorded past, we are able to state, in general terms, that the tendency of this early population was toward the rivers and deep swanrps, the rich valleys and the sea-coast. The physical inducements which impelled nomadic tribes to give a preference to such seats are obvious. Seldom are earth-mounds found at a considerable remove from water-courses. "Water and game were the chief attractions in the choice of a set tlement, liieh alluvial lands, whose fertility would make amends for the rude cultivation bestowed, upon them, were often selected as the sites of their vil lages. In those early days the rivers abounded with fish, and the deep swamps were replete with terrapins, alligators, deer, and other game. In the depths of these swamps, beneath the shadows of moss-covered trees and by the sides of the sluggish lagoons, large mounds are not infrequent. It is upon the islands,
ANCIENT TUMULI IN GEOKGIA.
129
however, and along the headlands of the coast, that they appear in greatest numbers.
Take, for example, that group of more than forty mounds upon the Colonel's Island, in Liberty County, located in the vicinity of a large spring, which for un numbered years has been sending forth its copious and refreshing waters. Besides the regular sepulchral tumuli composed of sand, the adjacent fields are liter ally hoary with shell-mounds and the debris of longcontinued encampments. Extended oyster-beds, neigh boring creeks abounding with crabs, shrimp, and salt water fish of every variety native to the coast, woods in former years well stocked "with game, the natural advantages of a high, dry bluff sheltered from north-* easterly gales, and this never-failing supply of fresh water, without doubt rendered this a very attractive spot to the Indian. His settlement here was perma nent and extensive. Most of the tumuli in this neigh borhood are sepulchral in their character. Such is the distinguishing peculiarity of nearly all the coast mounds.
The ancient tumuli still extant within the geo graphical limits of Georgia are frequently associated in groups, and at other times exist as isolated monu ments erected upon or near localities possessing some natural advantages for observation, defence, or for the facile procurement of food. In form they are circular, elliptical, quadrangular, and polygonal. Some are flat on the top, resembling truncated pyramids and trun cated cones. The prevailing type, however, is that of the conical earth-mound. There is every variety in size, from the large temple-mound on the Etowah-- more than sixty-five feet high, and with a summit diameter of over two hundred feet--to the small sepul-
130
ANTIQUITIES OF THE SOUTHERN INDIANS.
chral tumulus whose existence can scarcely be recog nized. Many are almost level -with the ground, and decomposing human, bones, mingled with fragments of pottery, He exposed upon the surface. Constructed of loose mould, clay, and sand, they are liable to constant diminution in size, and eventually to total obliteration. The consequence is, they are all more or less reduced, and we may readily believe that many of the smaller ones and those of oldest dates have entirely disap peared.
Aside from the careful and laborious preparation of their Chunky-Yards,' the construction of elevated foundations for their rotundas, and the erection of occasional and small tuniuli above some deceased per sons of note, it would appear that the Georgia tribes had well nigh abandoned the custom of mound-build ing prior to tho advent of the Europeans. In Plate XL. of the " Brevis Narratio" a we have a spirited representation of the ceremonies observed by the Florida Indians upon the occasion of the sepulture of their kings and priests. Located in the vicinity of tho village appears a small conical mound surmounted by the shell drinking-cup of the deceased, and sur rounded by a row of arrows stuck in the ground. Gathered in a circle about this sepulchral tumulus the berea-ved members of the tribe, upon bended knees, are bewailing the death of him in whose honor this grave-mound had been heaped up.
Bartram " commemorates the fact that in his day the Choctaws covered the pyramid of coffins, taken
1 See Bartram's " Creek and Cherokee Indians." ican Ethnological Society," vol. iii., part 1, p. 52.
3 Francoforti ad Mo3imm, De Bry. Anno 1591. 3 "Travels," etc., pp. 514, 515. London, 1792.
" Transactions of the Amer
MOUND--BUILDING WITIIIX THE HISTORIC PERIOD. 131
from the bone-hoiise, with earth, thns raising " a coni cal hill or mount."
Tomo-chi-chi pointed out to General Ogletliorpe a large conical mound near Savannah, in which he said the Yamacraw chief "was interred, who had, many years before, entertained a great white man with a red beard, -who entered the Savannah Kiver in a large vessel, and in his barge came np to Yamacrawr bluff.1
Within tlie range of my personal observation, glass beads, silver ornaments, hawk-bells, metallic ket tles, and occasionally a rusty gun or rifle-barrel, have been fonnd in earth mounds; but they evidently be longed to secondary interments, the graves in which they were located being either on the top or sides of the tumuli, and but a few feet deep.
Only in one instance has the writer discovered any article of European manufacture interred "with the dead in whose honor the mound was erected. Upon opening a small mound on the coast, a few miles below Savannah, an earthen pot, several arrow-heads, a stone celt, and a portion of an old-fashioned sword, were seen in immediate association with the decayed bones of a human skeleton. This tumulus was conical in form, seven feet high, and about twenty feet in di ameter at the base. It contained a single skeleton, and that lay, with the articles enumerated, at the bot tom and on a level with the plain. The oak handle, most of the guard, and about seven inches of the blade of the weapon still remained. The rest had perished from rust. Strange to say, the oak had more effectually than the metal resisted the " gnawing tooth of time." This mound liad never, prior to this occa-
3 "History of the Province of Georgia," etc., by John Gerai- "\Viili.\m DC Brahm, p. 38. Wormsloe, 1849.
132
ANTIQUITIES OF THE SOUTIIEKN INDIANS.
sion, been opened, or in any manner disturbed, except by tlie winds and rains of the changing seasons. The interment was primary, and the articles were lodged "with the dead before this mound-tomb was heaped above him.
It may be confidently asserted, therefore, that bur ial-mounds were erected by the Southern. Indians within the historic period; but it is not clear that the modern tribes had aught to do with the construction of those larger tumuli, in form resembling truncated pyramids and truncated cones, sometimes terraced, fre quently surrounded by a ditch or embankment, and intended for purposes other than those of sepulture.
AVhatever may have been the antecedent usages of the natives with respect to the erection of sepulchral tumuli, it is quite certain that their xise was discon tinued very shortly after the arrival of the colonists. Then, instead of being carefully disposed in the womb of the laboriously-constructed mound, the dead were exposed upon hastily-prepared scaffolds, hidden away in ledges of rocks, buried beneath the floors of their lodges, concealed in hollow trees, submerged in ponds, lakes and rivers, or interred in the forests with but ephemeral indicia, to mark their last resting-places. AVlieii used at all by the later tribes, these ancient tumuli seem to have been employed as convenient localities for what we may call secondary interments.
It is safe to assert that most of the mounds ante date the historic period. Compared with each other they differ materially in age. This is not to be won dered at, when we remember that the occupancy of this region by the red race, if we credit their traditions and properly interpret the monuments which they have left behind them, must have lasted
ANTIQUITY OF THE TUMULI.
133
for many generations. Some of these tumuli are not less than eight centuries old, while at least one, as we have already intimated, was thrown up after the European had. visited the New "World. In the absence of all definite information, the antiquity of these tumuli may be readily inferred from their lo cation, internal evidence, and from the growth of the forest-trees which overshadow them. One of the noblest specimens of the live-oak we have ever seen grew upon the summit, and with its majestic arms threw a protecting influence above and around the entire mound; the dead, nameless here for evermore ; his tomb a rude heap of native earth in the solitude of the wild-wood lie once loved so well; his com panions gone, his memory forgotten, and this pride of the forest seemingly a guardian of the consecrated spot, with its deep foliage affording an inviting retreat wherein the pleasant birds of spring might warble their morning and evening songs, its sturdy roots pre serving the symmetry of the grave, its overarching branches defending its yielding form from the ruthless influences of the tempest. Attired in its garb of sober green, with its drapery of sombre moss swaying slowly and solemnly in the ambient air, it appeared an aged mourner -watching over the dead of the chil dren of the forest. 1
If to the time probably consumed in the actual construction of some of the largest tumuli, -we add the period intervening between their completion and abandonment--the length of which, although entirely a matter of conjecture, could assuredly have been by no means inconsiderable--and then note the fact that,
1 This live-oak was nearly te-j feet in diameter, and we know that it is a tree of slow growth.
134
ANTIQUITIES OF THE SOUTHERN" IHDIAjSTS.
when iii'st observed by tlie whites, they were deserted and overgrown with forest-trees apparently as large as any whicli composed the surrounding forests--not forgetting the further circumstance, that the Indians who were domiciled here could impart to the inquir ing European not even a tradition of the time when or of the peoples by whom they were built--in endeav oring to ascertain their age, the mind is irresistibly led back to a remote date.
That the peoples who once possessed the hydrographical basin of the Mississippi, and, departing, left behind them, all along the banks of the Father of "Waters, in the valleys of the Ohio, the Scioto and else where, striking monuments of their labors, supersti tions, and combined industry, at some remote period occupied at least some of the fertile valleys of Cherokee, Middle and Western Georgia, is not improb able. The location and physical peculiarities of some tumuli and enclostires, the character of the remains found in and near them, the presence of stone idols and metallic ornaments, and the traditions of modern Indians--who regarded them with commingled igno rance and wonder--unite iii claiming for them not only a marked antiquity, but also a striking resem blance to the monuments of the Mississippi Valley. When compared with mounds which -we know to be the product of the labor of the ancestors of the pres ent Indians, characteristic differences are observed, for which we are sometimes at a loss satisfactorily to account.
While it may be regarded as a matter of specula tion whether the builders of the terraced mounds and enclosed works within the confines of Georgia were the actual progenitors of the Indians who occupied
MOUND--BUILDEES.
135
this country when it was first visited by the Europe an, and while we may not be able fully to explain how it came to pass that the later tribes were more nomadic in their habits, less patient of* labor, and so neglect ful of many of the customs which seemingly obtained among' the peoples "whose combined industry erected these enduring monuments--in the light of the Span ish, narratives, after a careful consideration of the rel ics themselves, and in view of all the facts which have thus far been disclosed, both by personal observation., and the investigation of others, while freely admitting that the modern Indians, from various causes, had ceased to engage in the erection of worlds in whose com:. pletion, with the indifferent implements at command, so much tedious physical effort was involved, we nev ertheless see no good reason for supposing that these more prominent tumuli and enclosures may not have been constructed in the olden time by peoples akin to and in the main by no means further advanced in semi-civilization than the red-men native here at the dawn, of the historic period. In a word, we do not concur in the opinion, so often expressed, that the mound-builders were a race distinct from and supeior in art, government, and religion, to the Southern Indians of the fifteenth and sixteenth centuries.
CHAPTER VI.
Mounils on the Etowah River.--Temple for Sun-worship.--Stone Images.--Fishpreserves.--Tumuli in the Valley of Little Shoulder-bone Creek.--Circular Enrth-work on the Head-waters of the Ogcoehee.--Stone Tumulus near Sparta. --Mounds on the Savannah Kiver.--Meeting between the Cacica of the Savan nah and Do Soto.
PASSING from these general observations, -we pro ceed to consider the physical peculiarities of some of the most interesting and prominent groups of ancient mounds and enclosures within the present geographical limits of Georgia.
The first ~we shall notice are located upon the right bank of the Etowah River, on the plantation of Colo nel Lewis Tumlin, a few miles from Oartersville, in Bartow County. Viewed as a whole, this group is the most remarkable within the confines of the State. These mounds are situated in the midst of a beautiful and fertile valley. They occupy a central position in an area of some fifty acres, bounded on the south and east by the Etowah River, and on the north and west by a large ditch or artificial canal, which at its lower end communicates directly with the river. This moat (Gr Gr, Plate I.) at present varies in depth from five to twenty-five feet, and in -width from twenty to seventyfive feet. No parapets or earth-walls appear upon its edges. Along its line are two reservoirs (D D), of
MOUNDS IN THE ETOWAII VALLEY.
137
about an acre each, possessing an average depth of not less than twenty feet, and its upper end expands into an artificial pond (I*), elliptical in form, and somewhat deeper than the excavations mentioned.
Within the enclosure formed by this moat and the river are seven mounds. Three of them are preeminent in size, the one designated in the accompanying plan (Plate I) })y the letter A, far surpassing the others both in its proportions and in the degree of interest which attaches to it.
To the eye of the observer, as it rests for the first time upon its towering form, it seems a monument of the past ages, venerable in its antiquity, solemn, silent, ; and yet not voiceless, a remarkable exhibition of the power and industry of a former race. "With its erection 'the modern hunter tribes, so far as our information extends, had naught to clo. Composed of earth, sim ple, yet impressive in form, it seems calculated for an almost endless duration. The soil, gravel, and smaller stones taken from the moat and the reservoirs were expended in the construction of these tumuli. The sur face of the ground, for a considerable distance around the northern bases, was then removed and placed upon their summits. "Viewed from the north, the valley dips toward the mounds, so that they appear to lift them selves from out a basin.
The central tumulus rises about sixty-five feet above the level of the valley. It is entirely artificial, consist ing wholly of the earth taken from the moat and the excavations, in connection -with the soil collected arouflcl its base. It has received no assistance whatever from any natural hill or elevation.
In general outline it may be regarded as quadran gular, if we disregard a slight angle to the south. That
133
ANTIQUITIES OF THE SOTTTHEIOT INDIANS.
taken into account, its form is pentagonal, with summit admeasurements as follows : length of northern side, one hxindred and fifty feet; length of eastern side, one hundred and sixty feet; length of southeastern side, one hundred feet; length of southern side, ninety feet, and length of "western side, one hundred feet. Meas ured ea,st and west, its longest apex diameter is two hundred and twenty-five feet; measured north and south, it falls a little short--being about two hundred and twenty feet. On its summit, this tumulus is nearly level. Shorn of the luxuriant vegetation, and tall foresttrees which at one time crowned it on every side, the outlines of this mound stand in bold relief. Its angles are still sharply defined. The established approach to the top is from the east. Its ascent -was accomplished through the intervention of terraces, rising one above the other--inclined planes leading from the one to the other. 1 These terraces are sixty-five feet in width, and extend from the mound toward the southeast. Near the eastern angle, a pathway leads to the top; but it does not appear to have been intended for very gen eral use. May it not have been designed for the priest hood alone, while, assembled upon the broad terraces, the worshippers gave solemn heed to the religious ceremonies performed upon the eastern summit of this ancient temple ?
East of this large central mound--and so near that their fianks meet and mingle--stands a smaller mound about thirty-five feet high, originally quadrangular, now nearly circular in form, and with a summit diame ter of one hundred feet. From its western slope is an easy and immediate communication -with the terraces of
MOUNDS IN" THE ETOWAH VALLEY.
139
tlie central tumulus. This mound is designated in the accompanying plate Toy the letter B. Two hun dred and fifty feet in a westerly direction from this mound, and distant some sixty feet in a southerly di rection from the central mound, is the third (C) and last of this immediate group. Pentagonal ill form, it possesses an altitude of twenty-three feet. It is uni formly level at the top, and its apex diameters, meas ured at right angles, were, respectively, ninety -two and sixty-eight feet.
East of this group, and within the enclosure, is a chain of four sepulchral mounds (^F F E E) ovoidal in shape. Little individual interest attaches to them. Nothing, aside from their location in the vicinity of these larger tumuli and their being within the area formed by the canal and the river, distinguishes them from numerous earth-mounds scattered lierc and there throughout the length and breadth of the Etowah and Oostenaula Valleys.
The artificial elevation E, lying northwest of the central group, is remarkable for its superficial area, and is completely surrounded by the moat which, at that point, divides with a view to its enclosure. The slope of the sides of these tumuli is just such as would be assumed by gradual accretions of earth successively deposited in small quantities from above.
The summits of these mounds, and the circumjacent valley for miles, have been completely denuded of the original growth which overspread them in rich profu sion. The consequence is, these remarkable remains can be readily and carefully noted.
We marvel at the amount of labor expended in their construction, and conjecture that they are either the product of the combined energies of a population
140
ANTIQUITIES OP THE SOUTHERN" INDIANS.
by no means inconsiderable, or else the representatives of the successive industry of perhaps several genera tions. Of one fact we may be persuaded, that there was not, in the eighteenth century, a single Indian tribe in this vicinity possessing either the disposition or the means of subsistence sufficient to enable it to apply the unprodxictive labor necessary for the erec tion of such works. !N~or were the Cherokees in such a social or political status as would have empowered their chiefs to have compelled such an expenditure of the physical energies of their nations. Nomadic tribes, relying upon the bow and arrow for subsist ence, and changing their seats under the influences of want and inclination, are loath to assume the erec tion of such huge earth-worts. "VVe have the positive testimony of the Cherokees, that they had not even a tradition of the race by whom these tumuli had been reared. During the period of our acquaintance with them idol-worship did not exist among the Chero kees ; and yet "within this enclosure three stone idols have been found, and numerous terra-cotta images fashioned after the similitude of man, beast, and bird. Of these stone idols it may be remarked, in passing, that two -were cut from a dark sandstone, were respec tively twelve and fifteen inches in height, and repre sented the male human figure in a sitting posture-- the knees drawn up almost upon a level "with the chin, the hands resting upon either knee. The third, and the most carefully-sculptured Indian idol the writer has ever seen, was a female figure made of a dark talcose slate. As, in a subsequent chapter, our atten tion will be specially directed to a somewhat careful examination of these and kindred antique images, and also to an inquiry into the nature and extent of idol-
STONE--IMAGES.
141
worship as practised by the Southern Indians at a re mote period, a more extended notice of these interest ing relics is here pretermitted.
Outliving the generations during which they were fashioned and perhaps invested with supernatural pow ers, and surviving the incoming and the outgoing of subsequent nomadic tribes, these stone images pre serve the peculiar forms and expressions which were in that age of shadows traced by the hand of semi-civil ized art upon the shapeless stone, and declare the former existence of peoples whose names are unknown, whose origin is the subject of mere conjecture, and whose history and customs are perpetuated simply "by a few scattered remains which, in the deluge of time, like floating plants have escaped the general shipwreck.
Unique specimens of idol-pipes, stone plates, large shell ornaments, and other relics not common among the Cherokees, confirm the impression that these tumuli were not the results of the labor of the modern In dians. The large trees which grew upon these mounds when they were first visited by the early settlers, and their utterly abandoned condition at the period of our primal acquaintance with them, add forcible testi mony in behalf of their decided antiquity. The great age of these structures is further demonstrated by the character of the works themselves, which are not the hastily-erected monuments of migrating bands, but the ruins of temples, areas, and burial-places, carefully considered, of massive dimensions, and indicating the consecutive, combined, and extensive labor of a consid erable population permanently established.
The eastern angle of the central mound is very prominent, and the upper surface in that direction is
14^
ANTIQUITIES OF THE SOUTHERN INDIANS.
more elevated. Just here have been found traces of hearths or altars, giving ample token of the continued presence of fire and perhaps of sacrifice. The terraces lie toward the east, and there is that about this tumu lus which induces the belief that it "was erected for religious purposes, and that upon its eastern summit religious rites were performed and oblations offered to the great divinity, the sun. The broad terraces and the adjacent dependent tumuli aiforded space for the assembling of worshippers at the appointed hour, when, from the elevated eastern summit of the large tumulus, the eye of the officiating priest caught the earliest rays of the rising sun, as, lifting his face from out the shadows of the distant hills, he smiled upon this beautiful valley.
In the true relation of the vicissitudes which at tended the Governor Don Hernando de Soto and soiue nobles of Portugal in the discovery of the province of Florida, 1 we are informed by the Gentleman of Elvas, that " on "Wednesday, the nineteenth day of June, the Governor entered jPacaha, and took quarters in the town where the Cacique was accustomed to reside. It was enclosed, and very large. In the towers and the palisade were many loopholes. There was much dry maize, and the new was in great quantity throughout the fields. At the distance of half a league to a league off, were large towns, all of them surrounded with stockades. Where the Governor stayed was a great lake near to the enclosure; and the water entered a ditch that well-nigh went round the town. Prom the River Grande to the lake was a canal, through -which the fish came into it, and where the Chief kept them, for his
1 Buckingham Smith's translation, pp. 112, 113. Bradford Club Series, No. 5. New York, 1866.
FISH--PRESERVES.
143
eating and pastime. With, nets that were found in the place, as many were taken as need required; and however much might be the casting, there was never any lack of them. . . . The Cacique of Casqui many times sent large presents of fish, shawls, and skins."
^Vhile the earth removed in the construction of the ditch and excavations was primarily employed in the erection of the tumuli within the enclosure, while they may in one sense be regarded as the sources of the mounds, and while their sizes and depths were, to a certain extent, regulated by the supply of material requisite for the completion of the projected truncated pyramid--"which we suppose to have been a temple-- and its dependent mounds, we are of opinion that, dur ing the progress of the entire work, direct reference was had to the final use of these excavations and of tliis canal as fish-preserves, whence the priests, ca ciques, and noted personages, of the nation, who prob ably dwelt "within the enclosure formed by the moat and the river, could at all seasons derive an abun dant supply of fish. The canal leading from the ar tificial pond in which it takes its rise, comnmnicates directly with both reservoirs, and, after passing them, empties into the Etowah. Through this canal fishes could have been readily introduced from the river into all three of these artificial lakes, and there propagated. Cane or wooden, wears---in such common use among the Southern Indians during the sixteenth century-- would have prevented all escape, and thus these reser voirs would have answered the purposes of FISH-PRE SERVES. Such we believe them to have been.
In the retired valley of Little Shoulder-Bone Creek, about nine miles from the village of Sparta, in Han cock County, may be seen another group of ancient
144
ANTIQUITIES OF THE SOUTHERN INDIAHS.
tumuli,1 not unlike tliose we have just examined. Of the date of tlieir construction, and of the peoples by whom they were erected, the Creeks and Cherokees professed utter ignorance. To the oft-repeated inquiry who "vvere the authors of these monuments and when were they built, the uniform response of the red-men "was, " We know not; our fathers found them here when they first possessed the land."
]From all these mounds the original forest-growth has been removed, and we are therefore denied the in formation "which "would be derived from an examina tion of the cortical layers of the venerable trees which formerly grew upon and overshadowed them after their abandonment by those to whose labors their ex istence was due. Here and there upon their summits still exist mouldering stumps and roots, affording ample proof of the vigor and proportions of that growth which the industry of a later race has carefully removed.
Approaching this series of tumuli from the west, the first which engages our attention (designated in the accompanying sketch by the letter G), in general out line nearly resembles a truncated cone; being slightly ovoidal, and with summit-diameters, measured east and west, and north and south, of, respectively, fiftytwo and forty-two feet. Its base-diameter, running east and west, is one hundred and forty feet. Meas ured at right angles, it falls a little short of this. Its present altitude is sixteen feet.
One hundred and fifty yards east of this mound is the largest tumulus of the group (C). It is a trun cated, pentagonal pyramid, its base-diameters, meas ured north and south, and east and "west, being respec-
T'la.te
TUMULI NEAB SPAETA, GEOBGIA.
145
tively one hundred and eighty and one hundred and eighty-four feet. Its summit-diameters, ascertained in the same directions, are respectively eighty and eightyeight feet. This mound is forty feet high.
By a reference to its profile (X), it will bo per ceived that it is higher toward the east. The approach to the summit was from the east, and the eastern third of the superior surface was not only elevated above the rest, but was also made scrupulously level. Here, a little below the surface, have been found traces of a hearth composed of baked clay or rude brick. Charred fragments of wood and other indications attest the former continued existence of iires upon this spot.
Considerable excavations have been made in the eastern slope. Composed, as it is, of the alluvial soil of the valley, the planters of the neighboring hills (entirely ignoring the claims of this ancient monu ment to preservation and respect--we had almost added veneration--at the hands of a utilitarian age), in by-gone years frequently resorted to it as a conven ient source of fertilization for their impoverished lauds.
..This tumulus, so august in its proportions, has in its construction derived no aid from any natural hill or elevation. It stands apart, and in the midst of a level valley. The slope of the sides is just such as would be assumed by the gradual accumulation of loose earth deposited from above.
It is not improbable that the Indians used the summit and sides of this tumulus for the purposes of
sepulture. Skeletons have been found near the surface, in a degree of preservation and possessing certain in dicia which forbid the belief that their inhumation was coeval with the construction of the mound.
The tumuli D, E, and F, appear to have been cle-
14(3
ABTIQtlTIES OF THE SOUTHERN INDIANS.
signed and used exclusively as burial-mounds. For so many years have they been traversed by tlie ploughsliare, and wasted by the winds and rains, that they liave doubtless lost mucli of their original proportions. Their surfaces are covered with fragments of human bones, and pottery, beads, arrow and spear heads, stone implements, stone ornaments, pipes, clay images, etc., etc.
The mounds C, D, and E, are isolated by a moat or ditch, indicated by the letters B B. The total area thus enclosed is between four and five acres. An additional ditch separates the mound E from the other two ; and, at the point H, are traces of an exca vation or reservoir, from which a third ditch. (K) leads to an adjacent small creek or stream emptying into Little Shoulder-Bone Creek. The earth taken from these moats or ditches, and removed ill digging the reservoir, was expended in the erection of the tu muli. There are no indications of embankments along their edges. .A.11 trace of this moat will soon disap pear, and marked changes have already occurred with in the recollection of the older inhabitants.
Within the enclosure, stone idols:--similar in ap pearance to those found in the valley of the Etowah-- and clay images, resembling the human form in dis torted, shape and feature, and fashioned after the simili tude of beasts and birds, have been gathered.
The fact has been distinctly attested by early travel lers, that the Indians of this region never worshipped ' idols. We have the further testimony that they not only never manufactured these symbols of pagan wor ship, but emphatically disclaimed all knowledge of the people liy whom they were made. Who, then, were these moti lid-builders, and who the artificers that chis-
TUMULI WEAK SPAETA, GEOKGIA.
147
elled tliese rude stone images which did not fall down from Jupiter ?
Every indication suggests and encourages the "be lief that this locality was, for a long period of time, densely populated. The surface of the ground, uot only within the enclosure, but up and down the val ley for a considerable distance, is replete with various relics. They lie also, in considerable quantities, com mingled with human bones, in the sepulchral mounds. Few and unsatisfactory arc the memories which they suggest. Feeble indicia of general customs, they do little else than furnish physical proofs of the former existence of nameless peoples who, living without let ters, have left behind them no legacies to history.
The surface of the enclosure--saving the presence of the mounds--is very level, and from it have been carefully removed all stones, bowlders, and fragments of rock, with "which other portions of the valley and the adjoining hill-sides abound.
On Plunkett Creek, abotit three-quarters of a mile distant, is a mound twelve feet high, with a summitdiameter of forty feet and a base-diameter of one hun dred and twenty-five feet. It is conical in shape, and its principal elevation is toward the east. Unlike the other tumuli in this valley, the present mound is mainly composed of fragments of rocks and stones ; and, apart from this fact, possesses no distinguishingpeculiarity. Its profile is shown in Fig. 2, Plate II.
Intermediate between this mound and the group which we have been considering, is an enclosed work, parallelogrammic in outline, containing an acre and a quarter. The ditch surrounding it is some ftrar feet wide, and between three and four feet deep. (^See Fig.
3, Plate II.)
143
ANTIQUITIES OF THE SOUTHERN INDIANS.
TJpon tlie head-waters of tlie Great Ogeechee Kiver, five miles from Sparta, is an earth-work, circular in form, with a gate or opening terminating at the creek. The embankment is still nearly three feet high, ancl upon it are growing trees as large and, to all appear ances, as old as any in the surrounding forest. (See Fig. 4, Plate II.)
The belief is current in the neighborhood, that this work was an intrenched camp of De Soto, but there is no satisfactory foundation for this impression.
Located upon a high, rugged ridge, three miles from Sparta, and in a direction opposite to that which led us to the so-called " Spanish Fort," are the remains of a stone tumulus originally fifteen feet high, and twelve feet in diameter at its base, nearly resembling a sugar-loaf in form. It was composed exclusively of fragments of rocks, carefully piled one alaove the other. A few years since a planter, moved by curiosity, un dertook the removal of this mound. The labor was but partially accomplished, and the only result attained was the almost total demolition, of this unique little tumulus. (See Fig. 5, Plate II.)
Tradition designates " Silver Bluif," or its vicinity, as the site of the ancient village of Cutifachiqui. There, if we rightly interpret the geography of the Fidalgo of Elvas, dwelt an Indian queen, young and attractive, who "with royal hospitality -welcomed to her capital and the freedom of hei* nation the adventu rous De Soto and his daring companions, lone wander ing and yet not lost amid the unbroken forests and howling wildernesses of a vast region hitherto un trodden by the white man. The historian of the ex pedition dwells at length and with evident satisfaction upon the reception extended by this Indian queen to
THE CACICA OP CUTIFACIIIQ1TI.
149
the knightly Ferdinand. Learning from three captives that a "woman held the sovereignty of this country, the General sent forward special messengers to her with offers of friendship. Her response of welcome was returned by her sister in person. Shortly after ward the queen appeared in a stately canoe, with an awning in the poop supported by a lance. She sat upon two cushions, and was accompanied by a number of Indian women--her attendants and maids of honor. Many escorting canoes followed. Invested with all the pomp and dignity which the limited resources of her age and race could throw around her, she crossed the Savannah Hiver and approached the bank where the Spanish Cavalier waited to receive her. Respond ing with ease, grace, and fervor, to his handsome ad dress, she landed and conferred upon him many pres ents--among them a pearl necklace, the beads of which are particularly mentioned as of great value and re markable size. The next clay the expedition crossed the Savannah Kiver in canoes and on rafts, and found rest, food, and refreshment, in the wigwams and be neath the wide-spreading mulberry-trees of the chosen town, of the cacica.
"Upon the eve of his departure, De Soto arrested the queen and forced her to accompany him on his for ward march to Chiaha. For seven long clays was she compelled to travel on foot through a wretched coun try, and it was not until the eighth day that she suc ceeded in making her escape. During this unwilling journey with the Spaniards she is said to have carried a casket made of reeds, containing pearls of great value. These she preserved; and so apt did she prove in concealing herself within the shadows of her native forests, that she, completely eluded the pursuit of the
150
ANTIQUITIES Oil' THE SOUTHERN INDIANS.
Spaniards, who were most desirous of retaining her as a hostage for their safe conduct through the territories of the neighboring cassiques who rendered homage to her.' The narrative leaves her in the wilds of an unknown forest returning to her people and chosen abode; and it may be that one of the rude earthmounds which tower along its banks, designates the last resting-place of the beautiful, the hospitable, the ill-treated queen of the Savannah.
No storied urn or monumental bust, no epitaph deeply graven on enduring marble, no sepulchral col umn, perpetuates her memory or her greatness; and yet certain tumuli, sternly wrestling with, all-subduing time, lonely and voiceless in this generation, even now repeat the story of the Indian queen, whose cordial welcome of and generous hospitality to the adventu rous, travel-worn, stranger, were requited by unkindness, ingratitude, and dishonor.
In 1776, Mr. Bartram states that there were in this vicinity what he is pleased to denominate Indian conical mounts, terraces, and areas, and also the re mains or traces of fortresses which were supposed to be ancient camps of the Spaniards, who formerly fixed themselves at this place in the hope of finding silver.2
Four years afterward, to the local history of this region another chapter "was added, whose incidents, authentic in their character, furnish a bright illustra tion of those partisan adventures and patriotic exploits which not unfrequently signalized the conduct of the Southern campaign in the clays of the good and great General Greene.
.CAPTURE OP FOKT GALPHIN.
151
Tlie annual royal present to the Indians, consisting of powder, ball, small-arms, liquor, salt, blankets, and other articles of which the impoverished Continentals stood most sadly in need, was, in May, 1780, on de posit at Fort Gralphin, about twelve miles below Au gusta, on the north side of the Savannah Kiver, await ing distribution. Colonel Brown's force at Augusta had been reduced by the detail of two companies of infantry, detached to guard this present. They were at that time stationed in the Stockade Fort at this point. Made aware of this fact through the vigilance of his scouts, carefully concealing his movement, and leaving his artillery and the tired of his battalion be hind under command of Eaton, Lieutenant-Colonel Lee determined to press forward at once and secure these much-coveted supplies for the American camp. Mounting a detachment of infantry behind his dra goons, by forced marches and -without the knowledge of the enemy, on the 21st of May, 1780, he halted his panting squadrons beneath the pines which skirted the field in -which Fort Galphin was located. The day was excessively sultry, and men and animals were so oppressed by heat and overcome by thirst, that his lit tle column -was for the time incapable of further exer tion.
After a short rest, Colonel Lee directed his dis mounted militia to make, unobserved, the circuit of the fort, and to attack it from, a point opposite to that which lie then occupied. This strategy was invoked under the impression that the garrison would be drawn, from the fort in the pursuit of these few militiamen, and thus its capture, by a rapid assault under his im mediate supervision, insured beyond a question. As was expected, so soon as the militia debouched from
152 AKTiQurrrES OF THE SOUTHERN INDIA NS.
the woods, the garrison flew to arms, and, rushing from the fort, pursued the militia, who, at first resisting feebly, quickly retired--their retreat being covered by some cavalry previously disposed for that purpose. At this juncture a rapid advance under Captain Ru dolph "was ordered, and the assaulting column easily gained possession of the-fort.
In the language of the author of the " Memoirs of the "War in the Southern Department," " the garrison, with the valuable deposit in its safe keeping, gave a rich reward for our toils and sufferings." *
An old brick house still stands which witnessed the prowess of the gallant cavalry colonel and his brave troopers on that sultry May morning.
But it is not of this old brick house with its Revolu tionary memories, nor of the bright blade of the Vir ginia chieftain, that we speak. It is not our purpose to pursue the track of the Spanish expedition, or to recount the traditions of the locality. Our object is simply to chronicle the existence and perpetuate the recollection of the prominent physical peculiarities of a marked group of ancient tumuli resting upon the left bank of the Savannah Kiver, some twelve "or fifteen miles by water below the city of Augusta. Thirty-five years ago this group numbered six mounds, but the restless river, with recurring freshets, encroaching steadily upon the Carolina shore, has already rolled its turbid waters over two of them, while other two have so far yielded to the levelling influences of the ploughshare as to be almost entirely obliterated. Consequently but two re main, and they only in major part, one-third of each having been washed away by the current; and the day ~ is probably not far distant when tradition only will
' Vol. ii., p. 89, 0. ssq. PhiladelpWa, 1812.
T'la.te. Zff.
TUMULI GJS MASON'S PLANTATION,
153
designate the spot once memorable in the annals of a former race as the site of monuments of unusual size and interest.
These tumuli are located on Mason's Plantation, upon tlie very edge of the Savannah Kiver, and in the midst of the wide, deep swamp, which here on either bank stretches away for miles, exhibiting one uniform, level, alluvial surface. AVhat was once a mighty forest, grand and impenetrable in its majestic trees and tan gled brakes, is now a rich cornfield whose harvests have for many years with a yield of a hundred-fold rewarded, the toil of the intelligent husbandman. The surrounding space being thus denuded of its original growth, the tumuli loom, up in uninterrupted propor tions, -while from the river, which has wellnigli cut them in twain, the observer enjoys a most favorable opportunity, as presented by their perpendicular fronts, for closely examining their physical composition. Fresh ets have performed what it would have required long days of toil to have accomplished, and even then the woi'k would not have been done half so well. It is sad to realize, however, that these encroachments which at present bring hidden things to view, and enable the examiner to pursue his investigations with facility, are dooming the objects themselves to early and absolute annihilation. Some forest-trees, chiefly beech and lo cust, still crown the summits and flanks of these frag mentary mounds trembling upon the brink of the re morseless river.
The largest tumulus, designated in the accompany ing sketch by the letter A (Plate III.), rises thirtyseven feet above the plain, and forty-seven above the water-line as it existed at the date of this visit. JVleasured east and -west, its summit diameter -was fifty-eight
154
ANTIQUITIES OF THE SOUTHERN INDIANS.
feet, while, in consequence of the encroachment of the river, when measured in a northerly and southerly direction, it fell a little short of thirty-eight feet. Its base-diameter, ascertained in an easterly and westwardly direction, was one hundred and eighty-five feet. Although its outlines have been somew-hat marred by the whirling eddies of the river, as its swelling waters, in the spring of the year gathering marvellous volume and impetus, have again and again swept by, inundat ing the entire swamp-region, this tumulus may be truthfully described as a truncated cone--its sides sloping gently and evenly, and its apex surface level. If terraces ever existed, they are no longer apparent. The western flank of this mound was extended for a distance some twenty yards or more beyond the point where it would otherwise have terminated, respect be ing had to the configuration of the eastern and south ern slopes. .A-bout two feet below the present siirface of this extension is a continuous layer of charcoal, baked earth, ashes, broken pottery, shells, and bones. This layer is about twelve inches thick. So far as our examination extended--and it was but partial--the admixture of human bones was very slight--the bones, of which there were vast numbers, consisting of those of animals and birds native to this region. One is at a loss to explain the existence of this stratum of charcoal, ashes, shells, fragmentary pottery and bones, unless upon the hypothesis that it comprises the debris of a long-seated encampment or permanent abode of the aborigines upon this little bluff. This stratum can be traced along the water-front of the mound, as though it existed prior to its construction. The superincum bent mass of earth seems to have been heaped above it. Where it penetrates the tumulus, it is wellnigh coiiici-
TTJMTJ1I ON MASOVS PLANTATION.
155
dent with a prolongation of what was at tlie time tlie surface of tlie surrounding swamp.
The 'mound itself is composed of the alluvium of tlie adjacent field, which, is a micaceous clay, richly impregnated with, vegetable mould. No traces of in humation could be perceived, and the composition of the tumulus was homogeneous as far as ascertained.
It is earnestly hoped that some one will carefully note from time to time the encroachments of the river, as in all likelihood the central portions of this mound will soon be laid bare, and then, its contents, if any, will be fully disclosed. Thus will an opportunity be afforded for a most satisfactory examination.
One hundred and twenty-five feet due east of this large tumulus, is the smaller mound designated by the letter B. Its appearance, general outline and composi tion, are so nearly analogous to those of the larger mound, that a specific description, is scarcely neces sary. It may be remarked, however, that, possessing a base-diameter of one hundred and fourteen feet, it rises fifteen feet above the surface of the ground and twenty-five feet above the level of the river.1
It will be perceived by a reference to the accom panying sketch (Plate III.), that these tumuli were, in days long since numbered with an unrecorded past, isolated by a moat (C C), whose traces are still quite observable. The enclosed space--the river forming the northern boundary--contains a conjectured area of about eight acres. Commencing at the river, eastwardly of the smaller monud and distant from its flank some thirty yards, this ditch extends in a south erly direction until it merges into what now seems to
1 For profiles of these tumuli, see letters F and G, Plate III. The water-line is represented by H,
15G
ANTIQUITIES OF THE SOUTHERN" INDIANS.
be a natural lagoon (D). Following tins in a westwardly course, it finally leaves it, and thence runs almost due north to the river into which it empties at a remove of about eighty yards from the "western flank of the larger tumulus. Here the communication with the river is still perfect, but the upper mouth of this moat is now dry. It varies in width from twenty to forty feet, and is in some parts wider still.'
In all probability the earth removed in the con struction of this canal was devoted to the erection of these tumuli; and there are here and there in their vicinity physical evidences of the fact that the sur rounding soil contributed to their further elevation. Terra-cotta vases, pots and pans, arrow and spear heads, stone articles of use and ornament, mortars, pipes, and bone and shell beads, are found in the adjacent fields, but there lives not a tradition of the time when, and of the tribes by whom, these tumuli were built. Lonely, storm-beaten, freshet-torn, they stand nameless and without a history in this genera tion--silent, yet convincing illustrations of the ephem eral character of the nomadic races which for centuries peopled this entire region, and, departing, left behind them neither letters nor monuments of art--nothing save these rude earth-mounds and occasional relics to give assurance of their former existence.
In the twilight of what by-gone and unrecorded century were these tumuli built ? A\rheiice came, and who the peoples that lifted them from out the bosom of our common mother ? Served they as friendly ref uge in seasons of freshet and of storm ? "Were sacred fires ever kindled upon your summits and within this consecrated area ? Within your hidden depths do the
1 This also may have been a fish-preserve.
i
TUMOTI OS" MASCOT'S PLANTATION.
157
brave and honored of your generation sleep that sleep which knows no waking until the final trump shall summon alike the civilized and- the savage to the last award? Or are ye simple watch-towers, deserted of your sentinels--forts, abandoned of your defenders? We question, but there are no voices of the past in the ambient air. We search among these tombs, but they - bear no epitaphs. The sacred fires, if ever kin dled, were turned into ashes long ago, and naught but darkness is here. We gaze upon these monuments, but thejt are inscriptionless, and the Savannah rolling its swollen "waters about them will soon sweep even these mute earth-mounds out of existence. For a few short moments this tawny-hued river will grow more turbid with the dissolving mass of native clay, and then, borne away upon its bosom, and settling darkly in the depths of this swiftly-moving stream, nothing will evermore be seen of these august witnesses of the memorable meeting between the Spanish Adventurer and the Cacica of the Savannah.
CHAPTER VII.
Tumuli on the Ocmulgee River, opposite Macon.--Brown's Mount.--Mound on Messier's Plantation, in Early County.
OF the mounds on the left bank of the Ocmulgee River, opposite the city of Macon, the largest and most noteworthy (A, Plate IV.), lying farthest down the river, is located upon the summit of a natural hill, and occupies a commanding position. The earth of "which it is composed was gathered in the valley and con veyed to the top of the hill so as, in the end, to in crease its elevation by some forty-five or fifty feet. The summit diameters of this tumulus, measured north and south, and east and west, are respectively one hun dred and eighty and two hundred feet. On the west is an artificial plateau, still about eight feet high, seventy-two feet long and. ninety-three feet wide.- On the north and east are three spurs or elevated ap proaches, over which, as paths, the laborers, during the construction of the mound, carried their .burdens of sand and clay in cane baskets, and, by means of "which, "when the tumulus "was completed, ascent to its summit "was rendered more facile. It is not improba ble that this was a temple-mound, used by priests and devotees in their established worship of the sun.
TUMULI WEAR MACON", GEOEGIA.
159
One hundred feet north of this tumulus is a second mound (B) about ten feet high, elliptical in shape, with a summit-diameter, measured in the direction of the major axis, of one hundred and twenty-eight feet. Northwest of this mound and distant between three and four hundred 3Tards, is the third of the group (C), its outlines marred by the elements, and its northern slope carried away by the excavation for the new track of the Central Railway. It is still about forty feet high and is conical in form--its mean summit-diameter being about eighty-two feet. On its top is the decayed stump of a tree, more than five feet thick.
About four hundred yards in a northeasterly direc tion is the last tumulus of this series (D). In general characteristics it closely resembles the mound last 1 mentioned. These mounds are all flat, and may be described as truncated cones, with the exception of the temple-mound, which assimilates the form of an oc tagonal, truncated pyramid. The temple-mound was erected for religious purposes ; the others were heaped up, probably, in honor of the dead. In their vicinity the fields are filled, with sherds, shells of the pearlbearing unio, and fragments of articles of ancient do mestic economy. Upon the acclivity east of the cen tral mound are the manifest remains of an aboriginal settlement. Here, in. excavating for the new track of the Central Kailway, the workmen a short time since unearthed, a few feet below the surface, several skele tons, in connection with which were found beads of shell and porcelain, a part of a cliscoidal stone, several arrow and spear points, two stone celts, a clay pipe, an earthen pot, and. other matters of a primitive character fashioned for use or ornament.
This excavation for the line of the railway neces-
160
ANTIQUITIES OF THE SOUTHERN INDIANS.
sitated the removal of a considerable portion of the northern side of the central mound. In tie conduct of this work, the laborers, while cutting through the slope of the mound, and at a depth of perhaps three feet below the superior surface, exhumed several skulls, regular in outline and possessing the ordinary characteristics of American crania. Associated with these skeletons were stone implements--the handiwork of the red race--and Venetian beads and copper hawkbells acquired through commercial intercourse with the early traders and voyagers. The fact was patent that at least some of these inhumations had occurred subsequent to the period of primal contact between the European and the Indian.
Passing below these interments--which were evi dently secondary in their character--and arriving at the bottom of the,mound, a skull was obtained -which differed most essentially from those we have described as belonging to a later inhumation. It "was vastly older than those of the secondary interments, and had been artificially distorted to such an extent that the cerebel lum -was quite obliterated, while the front portion of the skull had not only been flattened but irregularly compressed, so as to cause an undue elevation and di vergence to the left.
For the purposes of comparison we have (in Plate IV.-A) figured two skulls, the first (1) being that of a modern Indian buried upon the side of the mound only a few feet below the surface; the other, the crani um of the primitive man in "whose honor the tumulus was constructed. Of this latter skull we have both a front and side view (Figs. 2 and 3, Plate IV.-A).
Among the relics found in the vicinity of this old, artificially-compressed skull, was a total absence of
PKi:>IAP*Y AXD SECONDARY IXTEKMEKTS.
1 Gl
European ornaments. Here we have an interesting demonstration of the fact that these ancient tumuli were, in turn, used by tribes who perhaps had noknowledge the one of the other. The ^flattened and dis torted skull belongs to the mound-building people to whose industry the erection of these tumuli is to be referred. It was in perpetuation and in honor of such primal sepulture that this mound was heaped up. In the course of time these sepulchral and temple structures, abandoned of their owners, passed into the hands of other and later red races, who buried their dead upon the superior surface and along the slopes of these ancient tumuli, having at the time, perchance, no personal acquaintance with, and frequently not even a distinct tradition of, the peoples to whose exertions these evidences of early constructive skill were attributable.
In the absence of letters and of recorded memories most easily does one wave of human life sweep over another, obliterating all former recollections save such as are lodged in. the womb of mounds, or preserved in the generous bosom of mother earth :
" Tlic very generations of the dead Are swept away, and tomb inherits tomb, Until the memory of an age is fled, And, btiried, sinks beneath its offspring's doom." '
The Creeks did not claim that these tumuli were erected by them. They declared that they were here when their ancestors first possessed themselves of the region. Who these flat-head mound-builders were, is matter for conjecture. It may be that they were a colony of the Natchez, journeying hither from their old habitat on the banks of the Mississippi. Certain
102
ANTIQUITIES OP THE SOITTHEKN INDIANS.
it is, that tliese tumuli antedate the traditions of the Creeks who were native here at the period of the Eng lish colonization.
Below these mounds--in the valley-lands of the Ocmulgee, upon Lamar's plantation--are several large tumuli. The presence of these mounds, and the nu merous relics scattered throughout the length and breadth of the valley for miles, afford ample testimony that this rich alluvial soil was once the seat of a numerous and, perhaps, permanent population. The debris of frequent encampments along the bluffs of the river prove that the aborigines, dtiring the lapsed centuries, congregated here in numbers for fishing and hunting; and old clearings in the valley give evidence that they supported themselves in part by the cultiva tion of maize.
The many umo-shells overlying the surface of the fields and intermingled with the refuse piles of former encampments, corroborate the fidelity of the Spanish narratives and furnish present physical assurance that the natives of this region carefully collected tliese shells that the animals which they contained might serve as food, and their valves, so iridescent with pearly nacre, afford material for the manufacture of beads, gorgets, and other ornaments. From them, also, were pearls obtained, which the Indians perforated with heated copper spindles that they might be strung and worn as necklaces, armlets, anklets, and about the shoulders and waist.
The presence of gorgets, made of marine shells, and numerous columns of the strombiis giyas, some in an imperfect condition, and others entirely finished and perforated longitudinally so that they could be used as pendants, attest the commerc3 which existed be-
ATTCIENT FORTIFICATION OS BEOWU? S MOUNT. JG3
tween the coast Indians and those occupying' the in terior.
"BKOWN'S MOUNT," situated on the line between" Bibb and Twiggs Counties, from its summit affords a iine view of the city of Macon, "while, from its western exposure, wliic!i is very precipitous, the eye ranges all over the Oemulgee Basin and across the country far away to the valley of the Flint River.
Following the natural conformation of the summit boundaries, and at some points retired a distance of twenty yards or more from the edge of the hill, are the remains of an old "wall--constructed of bowlders of rock, and earth--which encircled and fortified the entire top of the mount. About sixty acres, I am informed, are thus enclosed. Attendant upon the wall are traces of both an outside and an inside ditch, the former being originally about ten feet wide and four feet deep, and the latter some three feet "wide and between two and three feet deep. The earth removed in the construc tion of these ditches was used, in conjunction with the stone-bowlders, in building this -wall. "Within the rec ollection of persons still living, this wall was four feet high, and between four and five feet in thickness. It will be perceived that the height of the wall was practi cally increased by the depth of the interior ditch; so that the defenders standing in the ditch would be com pletely protected from the shafts of their assailants.
The defensive abilities of this circumvallation were augmented "by elevated platforms and lunettes con structed all along the line at intervals of about thirty yards. The interior dimensions of these lunettes may be expressed by ten feet in front and eight feet in depth. By this arrangement, at close intervals, the defenders were thrown in advance of the line; and,
164
ANTIQUITIES OF THE SOUTHERN INDIANS.
elevated upon platforms, were enabled not only to de liver a powerful direct fire, but also with their arrows and spears to eniilacle the main line, thereby securing a double advantage in case of attack, and affording material aid to those who were defending the Avail or curtains connecting these advanced works.
Ill some places the Avail lias become well-nigh ob literated ; at other points it is still quite distinct, and its entire circuit, as well as the outlines of the lunettes, can be traced all along the crest of the hill. Upon the Avail, trees are growing more than three feet in diam eter. This was, without doubt, the work of the redmen, and in ancient times constituted a fortified re treat. Similar structures exist within the limits of Georgia and in many portions of the United States. It will be remembered that, in the absence of any speedier mode of transmitting intelligence, the Indians signalled by means of fires kindled upon prominent points. Through their intervention the approach of danger was heralded, and the lurid warning quickly repeated until the members of the tribe, through all their abodes, were rapidly put upon the alert. Such is the location of Brown's Mount, and so abrupt and commanding its exposure oil the west, that signal-fires kindled there could be readily seen and interpreted even by the primitive dwellers upon the banks of Hint Hiver. From the side which looks toward. JVlacon kin dred warnings--cloudy pillars of smoke by day and bright flames by night--would quickly summon the warriors of the Upper Ocmulgee, and put those, who there inhabited, upon notice. Doubtless, during the forgotten past, this fortified hill answered important military uses in the conduct of the ever-recurring strifes which existed among the red-men.
ANCIENT? FORTIFICATION OS" BROWN'S MOTJKT. 165
The impression, entertained by some, that this circumvallatkm was the -work of De Soto and his follow ers, is erroneous.
AVithin tlie enclosure are tile traces of two small earth-mounds, and near tlie northeastern side is a pond or basin, elliptical in form, covering about a quarter of an acre. Of late years it has been drained, and at the time of my visit it contained no water. The statement was made that this was an artificial basin and thai its "bottom had been plastered with clay at some remote period, so as the more effectually to retain the rain-water which would, from time to time, accumulate in it. I had no means at command for making an examination, and testing tlie truth of this assertion. The pond was overgrown with trees, and filled with decayed leaves and loam. To all appearances, it seemed a natural res ervoir, although it may be that the natives originally made this excavation with a view to supplying them selves with "water in the event of a siege. The natural supply of this fluid, upon ordinary occasions, was prob ably derived from four springs issuing from the north ern, eastern, southern, and western faces of the hill-- in each instance, within not much more than fifty yards of the Avail. Indications still exist tending to establish the fact that the paths leading to at least some of these springs were protected by stone walls or partially-covered ways. The summit of this hill is well adapted to cultivation, and, in one locality, I ob served a circular depression, about forty feet in di ameter, which suggested the belief that it might be the former site of one of those semi-sunken public gran aries in use among the Southern Indians, of which the early historians have given us substantial descrip tions.
166
ANTIQUITIES OF THE SOUTHERN INDIANS.
In the first volume of Pickett's " History of Ala bama" ' may be found a sketch, and also a description of a group of mounds on the plantation of ]VTr. Messier, in Early County, Georgia. Both have been repro duced without variation, or subsequent verification, in "White's " Historical Collections of Greorgia." 2 The unusual, haycock appearance of these tumuli and the prim, sharply-defined circumvallation have always seemed extravagant, and encouraged the belief that the plan and accompanying explanation had been carelessly prepared, and were in the main erroneous. Impressed with this conviction, and being unable to make a per sonal examination of the locality, the author requested .Major .Tames Audley Maxwell--a well-known and skilful civil engineer--to visit the spot and favor him with an accurate survey of such ancient earth-works as were now to be seen. This he has kindly done, and to him are we indebted for the following plan and de scription conveying a correct idea of the present condi tion of these interesting evidences of early construc tive skill. It will be readily noted how widely the impressions of an intelligent engineer differ from the notions of a casual observer.
The MESSIER MOUND--so called because situated on the plantation of Mr. Messier, of Early County--is lo cated about twelve miles east of the Chattahoochee Riv er, and from the summit of a hill looks down upon the narrow valley of the Little Colomokee Creek. Crown ing the natural hill -with an artificial elevation of fiftyfive feet, from its top is afforded a commanding view of the surrounding country. In the vicinity of this tumu lus -aucl stretching away to the west, are seen the culti-
1 Page ItlS. Charleston, 1S5I. s Pa"c 425. New York, 1854.
IS EAELT COUNTY, GEOBGIA.
167
vated fields of Mr. Messier, while on the east, north, and south, are tile swamps of Colomokee and its tributaries, beautiful in the luxuriant and variegated foliage native to this semi-tropical region. The most facile approach to the mound is from the west, access from any other quarter being rendered difficult by natural obstacles not easily overcome. The Messier mound is not one of a group, but stands apart, prominent in size and marked in its physical peculiarities. Other tumuli exist in the vicinity, one of them near enough to appear on the scale of the accompanying map; but none of these smaller mounds differ in any essential respect from the numerous hemispherical heaps of earth erected as burial-places by the Indians who formerly inhabited Southwestern. Georgia and Southeastern Alabama. 1 Tradition, speaking through the mouths of the de scendants of early European settlers, declares that the modern Indians lived here in large numbers, and that, while claiming the smaller mounds as the last restingplaces of their noted dead, they regarded the great mound with commingled wonder, ignorance, and su perstition. This traditional testimony is confirmed by the presence of numerous arrow and spear heads, frag ments of pottery, pipes, and other relics of the skill and industry of the red race. AVhether viewed near by or from a distance, the large tumulus seems but a huge mass of foliage--the outlines of this earthwork being concealed by leafy terraces of huge trees cover ing the sides and growing along the slopes from base to summit. The top of the mound is a level plane, and was long since denuded of all vegetation for the purpose of cultivation. Beneath the trees a tangled undergrowth of vines, bushes, and briers, in inex tricable confusion, forms an inviting retreat for the
108
ANTIQUITIES OF THE SOOTHERS' IHDIAWS.
rabbit -and the rattlesnake. It is only at some risk, and with indefatigable industry that the exact form of this huge earthwork can be determined. "While it is not singular that cursory observers should have formed erroneous impressions on this point, it is re markable and "worthy of condemnation that the results of such careless examinations should have been public ly heralded as conveying proper impressions of this interesting monument.
The form of this mound is that of the frustum of a four-sided pyramid; the top surface a level .plane--a rectangular parallelogram--the north and south sides being each sixty-six feet in length, and the east and west sides each one hundred and fifty-six feet long. The base-plane is not precisely level, but declines somewhat from the north toward the south, so that the vertical height of the mound at the northeast and northwest corners is fifty-three feet, while the vertical height at the southeast and southwest corners is fiftyseven feet. The northern boundary of the base of this pyramid is one hundred and eighty-eight feet long--the southern boundary about one hundred and ninetyeight feet, while the eastern and western boundaries are each three hundred and twenty-four feet. The slope of the east, west, and south sides is about one and a quar ter to one--or steeper than the natural slope of earth --while the north side slopes rather more than one and a half to one, which is about the natural slope of the earth of which this mound is composed. The foregoing description, in connection with the map and profiles (see Plate V.), cannot fail to convey an accurate conception of the shape of this mound. It must be remembered, however, that no earthwork can be said to conform precisely to any mathematical figure. The angles are
J
~v-
6f)Q y Ci rc/
'^'t'&ZZbxtfZT'
K PHOTO-I ITHOSRA PHlC CO. NKfQSBOffNtS
MOUKD IN EARLY COUNTY, GEORGIA.
169
always more or less rounded, and the slopes and sur faces to a greater or less degree convex or concave. The form of this mound agrees as accurately with the description given as does that of any modern earthwork with, the shape prescribed by the civil engineer. The slopes are even more perfect than those of railway em bankments. The fact that they are steeper than the natural slope must be explained upon the hypothesis of superior construction--as by the thorough packing of the earth in successive, thin layers. The greatest de parture from mathematical conformity to the pyramid occurs at the angles, which are rounded by curves of from five to fifteen feet in length. This may have been the result of design rather than the effect of time. Along the northeastern and northwestern an gles the ascent to the top may most conveniently be made, but there are no indications of any special pro vision for this purpose. There are no terraces. The space contained between the south side of the mound and the moat--easily recognized upon the map by its resemblance, in form, to the segment of a circle--is not a terrace. However important the use to which this space may have been dedicated, it possesses no digni fied elevation, but apparently occupies the same level with the original surface upon which the mound was erected. It is said that long ago a cavernous open ing in the southern slope of the mound was visible opposite the centre of this segment-like space, but there is now no indication that such an opening ever existed.
This tumulus contains about seventy-five thousand, cubic yards of earth, and woulol weigh from ninety thousand to one hundred thousand tons. By means of modern appliances its erection could be compassed
170
ANTIQUITIES OF THE SOUTHEEH" INDIANS.
at a cost of some fifty thousand dollars, provided the eartli was taken from the excavations from which the ancient mound-builders obtained it. The industrious labor of one thousand savages, properly applied for the space of one year, would have accomplished the work "with the aid of baskets or even earthen-ware pots for the transfer of the earth. These figures may or may not. engender disappointment. "We naturally incline to the marvellous, and if the reader expresses siirprise, let him compare the result with the scanty means then at command--the efiect with the appar ently meagre cause.
"Wo turn now to the surroundings of this huge tumulus. On the west lies a level plain well suited for the wigwams and streets of an Indian village-- for play-grounds and fields footh for manly exercise and the cultivation of maize. On the remaining sides the ground descends toward the neighboring streams, but there is no abrupt declivity from the immediate edge of the mound. For a distance of two hundred feet on the south, four hundred on the east, and seven hundred on the north, the natural plain is interrupted only by artificial excavations. These are A- B the moat, BCD the ditch, and E F G the pit, from all of which earth was taken and used in the construction of the mound.
From B to C the ditch is remarkably regular in form, and will average twelve feet in depth, ten feet in width at the bottom, and thirty feet in width at the top. _A_t the point C this ditch, as desci'ibed, ceases ab ruptly, and here commences a small ditch only two feet deep--apparently a natural channel worn l}y de. scending rain-water--deepening and widening until it reaches the edge of the swamp at D. From B to C
MOTJKD IN EABLY COTJNTY, GEOEGIA.
171
this ditch, is clearly artificial: from C to D it is seem ingly-not so. We should not positively conclude, how ever, that this ditch did not originally extend to the creek. From C to D a large ditch would naturally, with the lapse of time, become smaller in consequence of the constant accumulation of sand and clay brought down by the water. From B to C 011 the contrary, the ditch would receive no water except such as fell into its open mouth, and would preserve its outlines. No indications remain suggesting' that this clitcli was
formerly a covered way. The moat--so called for -want of a better name--is
simply a prolongation of the ditch, from 15 to A in the form shown in the accompanying sketch (Plate V.). From B to A it becomes uniformly -wider and shallowl er. At B, it is ten feet deep; at A two feet deep; and, half-way between those points, its depth does not ex ceed six feet. It is not probable that its original form and depth have materially changed. The slopes are so gradual that midway between the points A and B a buggy and horse can be driven across. There is noth ing remarkable about the segment-shaped space lying between the moat and the mound except its regularity
of outline. The ditch and moat furnished earth sufficient to
raise the mound to an altitude of only one foot. The rest of the material used in its construction was taken from the great pit E F Gr, which, although not accu rately measured, seemed just large enough to have furnished the required quantity. Its area is about two acres, and its average depth twenty-five feet, with easy slopes on the side nearest the mound. At the point E, however, the descent is perpendicular, and here an immense circular well, sixty feet in diameter and forty
172
ANTIQUITIES OF TH1I SOUTHERN INDIANS.
feet deep, may still be seen, clearly defined in all its parts. The bottom of this well is fifteen feet Tbelow the bottom of the pit, so that when the water in the well rises above the level of the bottom of the pit it flows off through the pit toward the point Gr, where this artificial excavation connects with a natural gorge in which rises and flows a small stream of never-failing water. Tho original head of this gorge was at Gr, and the location of the pit was doubtless selected with a view to draining off, through this natural channel, the water which would, necessarily accumulate in the pit during the process of its construction and materially retard the prosecution of the work. For this purpose the excavation was commenced at Gr, and progressed toward E, the water passing off in a direction oppo site to that in which the labor proceeded.
It is said that there formerly existed, and still re main in the vicinity, lines of earthwork a mile in length, but the writer could find no one able to point them out. Consequently he has not attempted to locate them on the map. South of the mound, at and from in to n, along the steep slope of the hill, the surface of the ground has been washed into numerous gullies in Avhich may be found many fragments of human bones. These are exposed after every heavy rain, but they ai'e so old and in such a decayed condi tion that they soon crumble into dust. The prob ability is that the side of the hill in this direction 'was extensively used for the purposes of sepulture. Some years ago a well was dug from the top of the mound, passing along its centre, to the depth of fifty feet. This investigation was not undertaken in the interest of science, but with the hope of finding precious metals and valuable stones. Disappointed in their expecta-
MOUND IN EAELY COUNTY, GEORGIA.
ItS
tions, the workmen subsequently closed this opening ; and from them no useful information has been gathered, touching the contents and stratification of the tumulus.
A.rrpw and spear heads, stone axes, fragments of quartz--not native to this region--and. numerous sherds of earthen vessels, variously and fancifully or namented, lie scattered upon the surface of the ground, and are turned up by the ploughshare in every direction.
Before the writer visited this mound he had formed a theory with regard to the method of its construction, which a careful examination compelled him to reject. Had he enjoyed the honor of serving as engineer-iiichief to his Majesty the King of the Mound-builders he would have suggested the selection of a hill like that represented by the heavy bi'olsen line in Fig. 4, Plate V. The earth taken from the dotted areas on either side, and placed so as to form the truncated pyramid indicated, by the coiitiiraous line, would, have produced a mound as large as the Messier mound, at an expenditure of only one-tenth the labor. The Mes sier mound, has received, however, no assistance what ever from any such device. It is entirely artificial, and the suggestion is named in this connection sim ply because it may turn out, upon the examination of other large tumuli, that they may have been built after this fashion.
White-oaks--some of them more than nine feet in circumference--are growing upon the sides of this mound. Their annual rings were not counted, nor is it known how many generations of forest-trees may have lived and died upon this tutnulug, each giving its tribute of soil to the surface, since the date of its abandonment by those who compassed its erection. If any superior stratum of baked earth, or an}' traces
174
ANTIQUITIES OF THE SOTJTHEEN INDIANS.
of sacrificial altars once existed upon the summit, they are not now exposed to view. 1
In the opinion of the writer, the Messier mound was erected not for defensive ptvrposes, but as a temple for the solemnization of religious rites--probably for the "worship of the sun. The erection of earth-walls in the vicinity was designed to facilitate the security and defence of a locality upon which, so much labor had been expended by these primitive peoples. These are incidental, however, and subordinate to the primal object, which was the construction of this huge moundtemple. As a defensive work this tumulus is well located, althoxigh its position is much inferior to many others which might be suggested in the neighbor hood.
It appears probable that in ancient times there existed an underground communication between the well E and the mound. That -well contains water now, and in all likelihood has never been dry.
Evidently these ditches and excavations were origi nally the sources whence was procured earth required for the erection of the mound. To what secondary uses they may subsequently have been dedicated must remain a matter of conjecture. In the religious festi vals of these primitive peoples ablutions subserved an
* In the description of this mound furnished by Dr. Charles A. Woodruff to Mr. Piekctt, more than twenty years ago, and published by him in his history of Alabama, hearth-stones are mentioned on the summit, with fragments of charred wood about them. These may have been, indicative ot'saciiftcialuses, or they may have been simply the places where the Creeks in latei- years kindled their signal-fires or eooked their daily food. The forest-trees then growing upon the mound were stated by Dr. Woodruff to be from four to five hundred years old. Of the earthwall enclosing the mound, Major Maxwell found no trace. "The arched passage, three hundred yards in length, leading from the large mound to the creek, and probably intended to procure water for religious purposes," spoken of by Dv, Woodruff, was probably nothing more than the segment-shaped moat and ditch described by Major Maxwell.
MOUND IN EABIiY COtHSTY, OEOEGIA.
175
important part, and the convenient presence of water "was deemed essential. AVhat precise significance may have been attached to its conveyance, in a particular way, to the neighborhood of the temple-mound is nowunknown. In the event of an attack, a liberal supply of this indispensable fluid was absolutely necessary; and it may be that in the location of the large reser voir and of the moat, respect was had to this contingen cy likely to occur at any moment in view of the preda tory habits of many of the tribes which, at that remote period, migrating hither and thither, sought to dis possess present owners of chosen seats which pleased their rude fancies or seemed most prodigal of those stores upon which they mainly depended for subsist ence. We conclude with one other suggestion, and it is this, that the large excavation and the semicircular moat may have been used as fisJirpreserves. We have already noted the fact that the Southern Indians, in the olden time, were in the habit of breeding fishes in artificial ponds, capturing them, with nets of their own manufacture as occasion required.
We might multiply examples, for they exist in vai'ious localities, but enough has, we trust, already been said, to convey a correct impression of the dis tinguishing characteristics of the ancient tumuli be longing to the class to which our attention has been directed.
Upon even a cursory examination of these groups of mounds with their attendant ditches, earth-walls, and fish-preserves, it is difficult to resist the impression that they are the remains of peoples more patient -of labor and in some respects superior to the nomadic tribes which, within the memory of the whites, clung around and devoted to secondary uses these long-
176
ANTIQUITIES OF THE SOUTHHEK INDIANS.
deserted monuments. There is not a considerable stream within the limits of Georgia in whose valleys tumuli of this sort are not to be found. They appealin Florida, and are very frequent iu Alabama, where truncated pyramids are even more abundant. Ten nessee, South Carolina, Mississippi, and Louisiana, are dotted with interesting monuments of tliis class. The occupation of this entire region by these mound-build-, ing peoples was by no means inconsiderable. It is in fertile valleys and upon the alluvial river-nats, whose soil afforded ample scope for agricultural pursuits, that these tumuli are mainly seen. "Why the older Indian tribes should have erected monuments so much more substantial and imposing than those which were con structed by the modern Indians, it is difficult to an swer. The Cherokees and Creeks did not, in many things, equal the aborigines of the sixteenth century as described by the historians of the expeditions. Whence the cause of this evident decadence in in dustry, craft, and power ? Can it be that the burdens imposed, the desolations wrought, and the diseases introduced among the natives by the Spaniards, con tributed to this demoralization ? Time was, if we may fairly judge from the appearance and manifest uses of some of these more august tumuli and their attendant relics, when those who built and cared for them held a position at least somewhat in advance of the later Indian tribes. Forming permanent settle ments, they devoted themselves to agricultural pur suits, erected temples, fortified localities, worshipped the sun, possessed idols, wrought largely in stone, fashioned ornaments of foreign shells, and occasionally of gold, used copper implements, and were not entirely improvident of the future. Such was the fertility of
MOUND IN EABLT COUNTY, GEORGIA.
177
the localities most thickly peopled by them, so pleas ant the climate, and so abundant the supply of game, that these ancient settlers were in great measure relieved from that stern struggle which, among no madic tribes and under more inhospitable skies, con stitutes the great battle with Mature for life. \Vith but few temptations to wander, except as their num bers increased, they seemingly devoted much attention to establishing their temples, protecting their settle ments, and confirming their chosen seats. And. yet they were not exempt from the vicissitudes which have befallen greater and more civilized nations--re verses born of the cupidity and cruelty of strangers, losses and positive destruction encountered at the hands of despoiling barbarians. It may be that they were compelled to abandon their valley-homes in con sequence of the incursions of more warlike peoples.
Certain it is that the inroads of the Spaniards violently shocked this primitive population, imparting new ideas, interrupting established customs, overturn ing acknowledged government, impoverishing whole districts, engendering a sense of insecurity until that time unknown, causing marked changes, and entailing losses and demoralizations perhaps far more potent than we are inclined, at first thought, to believe.
CHAPTER VIII.
Chunky-Yards.--Elevated Spaces.--Mounds of Observation and Retreat,--Tumuli on Woolfolk's Plantation.--Sepulchral Tumuli.--Chieftain-Mounda.--Custom of burying Personal Property with the Dead.--Savannah owes a Monument to Tomo-chi-chi.--Pamily or Tribal Mounds.--Cremation.
RESPONDING- to certain inquiries (propounded in all likelihood "by Dr. B. S. Barton) touching his personal observation of the customs, government, and antiquities of the Creek and Cherokee Indians, Mr. "William Bartram furnished the following plan and description of the CHUNKY-YAKBS (see p. 179).
"The Chunky - Yard of the Creeks, so called by the traders, is a cut>i-form area (A) generally in the centre of the town--the Public Square (located upon the square eminence C) and the Rotunda or great winter Council-House (situated upon the mound B, nine or ten. feet high) standing at the two opposite corners. It is generally very extensive, es pecially in the large, old towns, is exactly level, and sunk two, sometimes three feet below the banks or terraces (b b b b) surrounding it, which are sometimes two, one above and behind the other, and are formed of earth cast out of the area at the time of its forma tion ; these banks or terraces serve the purposes of seats for the spectators. In the centre of the yard
CIICHKY--TAEDS.
there is a low circular mount or eminence (c), in the centre of which stands erect the chunky-pole, which is a high obelisk, or four-square pillar declining upwards to an obtuse point, in shape and proportion much re sembling tho ancient Egyptian obelisk. This is of
wood--the heart or inward resinous part of the sound pine-tree--and is very durable ; it is generally from thirty to forty feet high, and to the top of this is fast ened some object to slioot at with bows and arrows, the rifle, etc., at certain times appointed. Near each corner of the lower and further end of the yard stands erect a less pillar, or pole (d d), about twelve feet high: these are called the slave-posts, because to them are bound the captives condemned to be burnt, and these posts are usually decorated with the scalps of their slain enemies: the scalps, with the hair on them, and
ISO
ANTIQUITIES OF THE SOUTHERN INDIANS.
strained on a little hoop, usually five or six inches in "width, are suspended by a string six or seven inches in length round about the top of the pole, where they re main as long as they last. I have seen some that have been there so long as to lose all the hair, and the skin remaining white as parchment or paper. The pole is usually crowned with the white dry skull of an enemy. In some of these towns I have counted six or eight scalps fluttering on one pole in these yards. Thus it appears evidently enough that this area is de signed for a public place of exhibition of shows and games, and formerly some of the scenes were of the most tragical and barbarous nature, as torturing the miserable captives with fire in various ways, as causing or forcing them to run the gauntlet naked, chunked and beat almost to death with burning chunks and fire-brands, and at la.st burnt to ashes.
" I inquired of the traders for what reason this area was called the chunky-yard ; they were in gen eral, ignorant, yet they all seemed to agree in a lame story of its originating from its being the place where the Indians formerly put to death and tortured their captives, or from the Indian name for it, which bears such a signification.
" The Indians do not now (ITl'S--1789) torture their captives after that cruel manner as formerly; but there are some old traders who have been present at the bui'ning of captives.
" I observed no chunky-yards, chunky-pole, or slaveposts, in use in any of the Cherokee towns ; and when I have mentioned in my journal chunky-yards in the Cherokee country, it must be understood that I have seen the remains or vestiges of them in the ancient ruins of towns; for in the present Cherokee towns
CHUNKY--YATID3.
181
that I visited, though there were the ancient mounts and signs of the yard adjoining, yet the yard was either built upon or turned into a gai'den-spot, or the like.
" Indeed, I am convinced that the chunky-yards now or lately in use among the Creeks are of very ancient date--not the formation of the present Indians. But in most towns they are cleaned out and kept in re pair, being swept very clean every day, and the poles kept up and decorated in the manner I have men tioned." '
The physical traces of these chunky-yards are still extant in various portions of the State of Georgia. In the soutlrwestern part of the State the forms of these tumuli and enclosed areas, and their relative positions in association with the outlines of the general settle ment, are in some instances quite observable. There are also spaces, parallelogramic in shape, elevated from two to four feet above the surface of the ground, uni formly level at the top and free from irregularities, which apparently were designed as play-grounds. Some of these were rendered hard by an admixture of clay and would have afforded excellent opportunity for rolling the discoidal stones which contributed so largely to the amusement and gaming proclivities not only of the Southern, but also of many of the other North American Indians. ~VVe will have occasion, in a subsequent part of this work, to notice more particu larly the use of these discoidal stones.
In order to facilitate the rapid communication of intelligence, upon an emergency, the Southern In diana erected conical earth-mounds upon commanding
1 Transactions of the American Ethnological Society, vol. iii., part 1, pp. 3430, 51, 52. New York, 1833.
18-3
ANTIQUITIES OF THE SOUTHERN INDIANS.
points, such as the tops of hills, or elevated river-"bluffs. Fires kindled upon their summits could be readily recognized and interpreted. The signals thus given were repeated from the tops of kindred mounds within convenient distances; and so, in. the absence of the warning bugle-note, the sound of drum, the booming of cannon, and tlxe passage of the electric spark, within a short period an entire tribe could be put upon the alert. These MOUNDS OF OBSERVATION are recognized by their peculiar situations, and from the further fact that they contain nothing other than the traces of the fires once kindled upon them now underlying the roots of overshadowing trees.
Striking examples of this class of mounds ma}' be seen on Woolfolk's plantation, on the Chattahoochee Hiver a few miles below Columbus, and at other points along the line of that river. Some years since, one of the largest was used to construct a heavy dam, and nothing was found in it save a shell drinking-eup and bits of charcoal. These tumuli are located with direct reference to the facile transmission of signals along the reaches in the river, and are so disposed that fires kindled upon theB- summits may be readily seen from a distance, and repeated. Situated in the riverswamp--which is liable to annual overflow--they served as safe retreats for the natives during freshets. On various occasions have the field-hands and planta tion-animals sought refuge upon the summit of the large truncated mound which stands just in rear of the negro quarters on this valuable place. Many tumuli of a like character might be mentioned, but these will serve as examples.
It was the remark of Ulloa, " If we have seen one American, we raaj be said to have seen all, their color
CHIEFTAIN--MOUNDS.
183
and make are so nearly alike." So might we affirm, in a general way, of the SEruxcrmAL MONUMENTS of the Georgia tribes. Although assimilated by many ob vious resemblances, for the purposes of our present description, they may be considered as resolving them selves into one or the other of the following classes.
Tumuli containing a single skeleton, or at most two or three skeletons, "we designate CIIIEFTAIXMOUNDS. The erection of such tumuli by the Florida Indians in honor of their deceased caciques and priests, is mentioned in the " Brevis Narratio." * Such mounds, varying in height from five to twenty-five feet, are found in many localities, and usually occupy promi nent positions in the vicinity of the spot which con stituted the village-site. They are for the most part conical in form, and the human bones which they con tain do not indicate the action of fire. "JSTot infre quently, the dead was interred in a sitting posture.
Such "was the case in a large mound carefully opened by the writer upon the Colonel's Island. The corpse had evidently been placed upon the ground and held in position while the loose sand was heaped around and above. lu the neighborhood of the feet and hands were numerous bone and shell-beads "which, at the time of the inhumation, encircled the "wrists, arms and an kles. Near the skeleton lay three stone axes, several spear and arrow heads, two pipes of rather unusual size--one of clay and the other of steatite--and a terra-cotta bowl, the property of the deceased at the pe riod of.his death.
In another mound the. body had first been seated in the centre of the spot to be surmounted by the tumu lus, and there, with his possessions deposited by his
1 Plate xl. Francoforti ad Momum. De Ely, anno 1591.
184
ANTIQUITIES OF THE SOUTHERN INDIANS.
side, was securely encased m a covering1 of tenacious red clay, six or eight inches in thickness, and ovenshaped. In this manner--the clay "becoming dry and hard--the sitting posture was maintained while the earth-tomb was heaped above.
Sometimes a stout light-wood post was first driven into the ground, and the dead, seated with their backs to the post, were securely lashed to it by means of thongs or grape-vines. Two instances of this sort have been brought to our knowledge. In one mound a single skeleton was found at the foot of the post. In the other the remains of three skeletons appeared, back to back, the post being in the centre.
Captain Bossu * informs us that the Alibamons buried their dead in a sitting posture, stating, in justification of the custom, that rnan being upright, should have his head turned toward heaven, which was to be his habitation. "They give to them," he con tinues, " a calumet and some tobacco to smoke, that they may make peace with the inhabitants of the other world. If the corpse be of a warrior, he is buried with his arms, which are a musket, some powder and bullets, a quiver full of arrows, a bow, and a hatchet, or club ; and besides these a mirror and some vermilion, with which they may dress themselves in the other world."
Upright burials are said, by Surveyor-General I-iawson,a to have been practised by the Carolina In dians.
In preparing their dead for sepulture, the Muscogulges placed the corpses in a sitting posture, deposit ing writh them such articles of property as "were held of greatest value." In celebrating the funeral rites of
' " Travels through Louisiana," etc., vol. i., p. 257. London, 1771. 1 " History of Carolina," e-e., p. 182. London, 1714. Bertram's "Travels," etc., p. 618. London, 179-!.
PERSONAL PBOPERTY BUKIED WITH THE DEAD. 185
a chieftain, the Cherokees seated the corpse in the torn!) "with the face turned toward the east, the head anointed with bear's oil, and the countenance painted red. He was attired in his finest apparel, " having his gnu and pouch, and trusty hiccory bow, with a young panther's skin, full of arrows, alongside of him, and every other useful thing he had been pos sessed of." '
The practice of depositing ill the grave all articles which the deceased deemed most valuable, obtained among all the Southern tribes. It has been truthfully remarked that " in all ages when the disengaged ac tivity of man. ever carries a keen and military edge with it, and his great employment is necessarily -war and the chase, the weapons of both would naturally be deposited "with the dead."
The ancient Germans contributed to the funeral pile the arms and the horse of the deceased. A.mong the more civilized Grecians expensive vases, mirrors, and ornaments were lodged in the tombs of the de parted. The grave has often proved the receptacle of treasure, and the storehouse of all that was most valuable among the possessions of the deceased. The souls of the Scythian kings and the Peruvian Incas -were, by costly immolations, richly furnished forth with companions, the most select, for the other wise lonely journey. Even the sepulchre of David was made the thesaurus of more than three thousand, talents.' In a strange land this custom was not neg lected by the Indian. During the visit of Tomo-chichi to England, in 1734, one of his companions died in London, of the small-pox. Previous to interment in
1 Adair'3 "History of the American Indians, 1' p. 182. London, 1775. a Squier's " Antiquities of Xew York," p. 114. Buffalo, 1851.
ISO
ANTIQUITIES OF THE SOUTHERN INDIANS.
the church-yard of St. John's, "Westminster, the body -was .sewn up in a blanket and bound between two boards. The clothes of the deceased, with a quantity of glass beads, some pieces of silver, and other articles of personal property, were thrown into the grave and buried with him.
The Spaniards under De Soto obtained large quantities of pearls by riiling the tombs, and pil laging the temples in which dead Indian chiefs were lying in state. The later graves which the Cherokees have left in Northern Georgia assure us of the fact that this custom of depositing with the dead all articles of value, the property of the deceased, was observed long after the establishment of commercial intercourse be tween the Indians and the Europeans. "We are also advised that these deposits were held sacred, and that among these tribes the graves of the departed, no matter how rich in coveted treasure they might have been, were never rifled.
It is to the graves of the common dead and. the tumuli erected in honor of departed chieftains., priests, and distinguished warriors, that we are largely indebted for many of the most interesting and perfect relics which grace our collections, and acquaint us with the condition of the arts among these primitive peoples.
Returning from, this digression, we may assert that these chieftain-mounds, when once completed, were never reopened for the reception of other bodies. The fact that, as a general rule, only a single skeleton is found in these mounds, and the further circumstance of their prominent size and location, very properly, we think, designate them as the last resting-places of the chiefs or distinguished personages of the tribe. Upon this supposition we are enabled the more readily to
CHIEFTAIN--MOUKDS.
187
understand the secret of their superior proportions. They may be regarded as the offering of the tribe or community--each member with ready hand assisting in erecting over the deceased leader a mound which, while it perpetuated the name and deeds of the honored dead and remained a monument of tribal respect and gratitude, begat also a pleasing satisfaction in the breast of all who had aided in its construction. Each of these silent, wasted mounds had its legends transmitted from sire to son, its heroic memories which brought the warm blood of conscious pride to the cheek alike of warrior and maiden ; but they have all perished "with those whose delight it was to perpetuate them.
These chieftain or priest mounds may be considered as individual in their character, the result of one im pulse, the consummation of a general labor prosecuted without intermission to completion. When we affirm that when once finished they were never reopened to admit the sepulture of parties other than those in whose honor they were erected, we take no note of those secondary interments, frequently occurring upon the tops and sides, which were probably made by later peoples, strangers to the original and distinctive memories of these tumuli.
Composed of sand, clay, mould, and sometimes of shells, the slope of their sides is such as would be assumed by the gradual accumulation of loose material piled from a\>ove. Often pits and sunken spaces in the immediate neighborhood indicate the localities whence was obtained the earth .expended in their construction.
These primitive peoples were at one time careful in the erection of marked tumuli above deceased kings
1S8
ASTIQUITIE3 OF THE SOUTHERN INDIANS
and priests. "With inferior means at command, they perpetuated "by physical signs the memories of the places where they slept with far greater zeal than did the Europeans the graves of the greatest of the redmen who had proved themselves invaluable allies, and through whose influence an infant colony was pre served, in the midst of a howling wilderness, from cruel attack and absolute massacre.
^Near Yamacraw bluff--a spot rendered memorable by the landing of General Oglethorpe, the founder of the colony of Georgia--Tomo-chi-chi, Mico of the Yamacraws, extended the open hand of welcome to the dis tinguished stranger, and tool? upon himself vows of friendship which he never ceased to observe until he bowed, his hoary head in death not far from the ancient pines beneath whose hospitable shade the governor first pitched his tent. In the presence of the colonists, few, feeble and filled "with doubts and ap prehensions--before his followers brave and jealous of their moss-clad forests, in a manner at once expres sive of genuine hospitality and redolent of that im agination so characteristic of his race, he presented General Oglethorpe with a buffalo skin adorned with the head and feathers of an eagle. " The eagle," said he, " is an emblem of speed, and the buffalo of strength. The English are as swift as the bird and as powerful as the beast, since like the former they flew over the seas to the uttermost parts of the earth, and like the latter are so strong that nothing can withstand them." Wisely divining in this small band the seeds of a great nation whose superior intelligence and resources were destined to exercise a controlling influence over his people, he added; " The features of the eagle are soft and signify love; the buffalo-skin is warm and denotes
BT7KIAL OF TOMO-CHI--CJII.
189
protection; therefore I hope the English will love and protect the little families of the sons of the forest."
Firm in his friendship, even unto the end, at the advanced age of ninety-seven he breathed his last; and, dying, desired that his body might be interred among his friends, the English, in Savannah. This request was complied with, and he was bulled with military honors in Percival Square.'
It may appropriately be asked, Where is his monument ? Over this mico, the white men--those whom he counselled, assisted, and saved, aud their descendants--have reared not even a simple moundtomb. To them did he confide the solemnization of his funeral rites and the perpetuation of his last resting-place, and they have paid no tribute to the memory of his grave. Of such neglect, think we, would not they have been guilty whose primal wrath against the early colonists was, through the persua sions and influence of this aged mico, turned into friendship.
To herself and the recollections of her infant clays, to the expressed wishes of General Oglethorpe who purposed the erection of a suitable shaft in honor of this departed king, as an honest acknowledgment of the debt for which she stands bound to her first aud best friend among the red-men, does Savannah owe a fitting monument to the brave, the generous, the noble-hearted Indian chief, the venerable Tomo-chi-chi.
Tumuli filled with, numerous skeletons may be regarded as FAMILY or TRIBAL MOULDS. The Indians of Southern Georgia frequently burnt their dead.
1 C. C. Jones, Jr. "Historical Sketch of Tomo-cbi-chi," pp. 120-127. Albany, N. Y., 1868.
" Plan of the City of Savannah and its Fortification, by John Gcrar William Da Brahni," p. 36 of " History of the Province of Georgia," etc. "VVormsloc, 1849.
190
ANTIQUITIES OF THE SOUTHERN" INDIANS.
This custom, however, was not universal, and it ob
tained to a very limited extent among the tribes resi
dent in the middle and upper portions of the State.
The practice of reserving the skeletons until they had
multiplied sufficiently to warrant a general cremation
or inhumation seems to have been adopted.
It was no easy task for the aborigines to erect a
tumulus. Hence, saving the construction of grave-
I
mounds in honor of distinguished personages, the labor
i
of sepulchral mound-building was postponed until the
/
accumulations of the bone-house claimed the attention
*'
of an entire community. Adair says that the bones of
-J-
those who died away from home or were slain in
I
battle were carefully preserved and, at some conven-
I
ient season, brought back and interred in a solemn
*
manner. To be deprived of the customary rites of
sepiilture was a calamity which an Indian could not
contemplate with indifference.
Funeral rites the Itomans called justa, the Greeks
Si'/caw, thereby intimating the inviolable obligation
which Nature imposed upon the living to perform the
obsequies of the dead. As among these civilized
nations the belief existed that the souls of the departed
could not be admitted into the Elysian fields unless
suitable funeral rites had been duly solemnized, in
like manner did the red-men cherish the faith that
a becoming observance of their rude obsequies was
essential to the entrance of their spirits into the hunt
ing-grounds of the blest. Here we have an explana
tion of the reason why they so carefully, in that remote
period, collected the skeletons of their dead and laid
them to rest in the burial-places of their kindred.
Bartram noticed among the Choctaws the follow
ing funeral custom: " As soon as a person is dead,
FIOTEBAL CUSTOMS OF THE CIIOCTAWS.
191
they erect a scaffold eighteen or twenty feet high, in a grove adjacent to the town, where they lay the corpse, lightly covered with a mantle. Here it is suffered to remain, visited and protected by the friends and relations, until the flesh becomes putrid, so as easily to part from the "bones; then undertakers, who make it their business, carefully strip the flesh from the bones, wash and cleanse them, and. when dry and purified by the air, having provided a curiouslywrought chest or coffin, fabricated of bones and splints, they place all the bones therein. It is then deposited in the bone-house--a building erected for that purpose in every town. Wnen this house is full, a general, solemn funeral takes place. The nearest kindred or friends of the deceased, on a day appointed, repair to the bone-house, take up the respective coffins, and, following one another in the order of seniority--the nearest relations and connexions attending their re spective corpse, and the multitude following after them--all, as one family, with united voice of alter nate Allelujah and lamentation, slowly proceed to the place of general interment, where they place the cof fins in order, forming a pyramid; and lastly cover all over with earth, -which raises a conical hill or mount." '
These observations of Mr. Bartram are fully cor roborated by the statements of Captain Bossu," Mr. _A.dair,a and others.
Upon the islands and lieadlancls along the coast, the skeletons, with a requisite amount of wood, were first placed in a pile upon the ground. Kire "was then applied, and, above the smouldering remains carelessly
1 " Travels through North and South Carolina, Georgia," etc., pp. 514, 515. London, 1792.
""Travels through Louisiana," etc., vol. i., pp. 298, 299. London, 1771. ' " History of the American Indians," pp. 183, ct scq. London, 1775.
192
ANTIQUITIES OP THE SOUTHERN INDIANS.
heaped together, a mound of earth was erected. The charred bones and partially-consumed fragments of wood are seldom seen until we have reached the level of the plain upon which the tumulus stands. W^ith rare ex ceptions, tribal mounds of this description contain Taut a single stratum of bones, showing that when .the cremation was ended and the tumulus finished, it was never reopened. As may well be expected, the bones in these mounds are disposed "without order. Being at best bnt fragmentary in their character, they are in termingled with ashes, charred pieces of wood, broken pottery, cracked pipes, and other relics sadly impaired lay the action of fire. The fires kindled in solemniza tion of these funeral customs were so intense as in some instances to crack the stone celts deposited with the dead. Shell ornaments entirely disappear, and the ordinary clay-pipes are generally broken to pieces.
La Hontan states that the natives dwelling upon the banks of the Mississippi burnt their dead, " reserv ing the bodies " until they had accumulated sufficiently to warrant the general burning, which "was performed out of the villages and in certain places set apart for that purpose. Du Pratz,1 on the contrary, asserts posi tively that "none of the nations of Louisiana were acquainted with the custom of "burning their dead." In the opinion of Mr. Haywood," some of the Tennes see mounds afford ample evidence of cremation.
-A_s we have already intimated, tumuli declaring un mistakably the fact that the skeletons -which they cover were burnt prior to the inhumation, are exceptional in their character; and, so far as our observation extends, are chiefly confined to the coast-region of the State.
1 " History of Louisiana,'1 vol. ii., p. 213. London, 1763. 3 " Natural and Aboriginal History of Tennessee," p. 138, Nashville, 182S.
BAKKOW IK LOW GKOUNDS OF THE BIVAKNA. 193
Why this custom should have obtained in some in stances, and not in others, -we are unable to explain.
Mr. Jefferson * examined, with considerable care, a barrow on the low grounds of the Kivanna, about two miles above its principal fork, opposite some hills on which there had been an Indian town. It proved to be a repository of the dead, and he conjectured that it contained not less than a thousand skeletons. In this mound the bones lay in strata, separated by interven ing spaces of earth, the skeletons of the different strata indicating the fact that they had lain for unequal pe riods in the ground--those nearest the surface being least decayed. The first collection of bones had been deposited on the ground and covered with stones and earth. A second had been laid on this, and covered in ' like manner. Other depositions were added from time to time, uutil the tumulus was completed.
Mounds have been opened by the writer, in various portions of Georgia, whose construction was compassed in a similar manner. Generally, however, these sepul chral tumuli contain but a single stratum of bones, and these laid upon the surface of the earth. The skeletons were deposited in a horizontal position, and were often piled one upon the other in such numbers, that the laj-er of bones, despite the weight of the superincum bent mass of earth, was sometimes a foot or more in thickness. In building these mounds the adjacent earth was used; and it would appear from numerous fragments of pottery, and from large mussel and conchshells intermingled with the soil constituting the tumuli, that the sand and clay were first scooped up by means of these shells, and then transported in terra-cotta ves sels, many of which were broken during the operation.
1 " Notes on Virginia," query xi.
194
ANTIQUITIES OF THE SOUTHEKN INDIANS.
These sherds and shells have no connection with the relics deposited with the dead. Mr. Haywood * sug gests, from personal observation, that similar means were employed in the erection of some of the "burialmounds of Tennessee. Baskets made of split cane and rushes were, doubtless, freely engaged in the convey ance of sand and other materials for the construction of these tumuli.
It is unnecessary to mention the particular locations of tumuli of this class, because they are still to be seen in nearly every part of the State. In form they are circular or elliptical, varying in height from two to twenty feet, and in. diameter from twenty to one hun dred and fifty feet. In cultivated fields many have been so sadly "worn away by the ploughshare and the action of the elements, that they are nearly level with the ground--fragments of bones and scattered relics lying exposed upon the surface.
1 " Natural and Aboriginal History of Tennessee," pp. 138, 139. Nashville, 1823.
CHAPTER, IX.
Shell-Mounds.--Tumulus on Stalling's Island.--Shell-Heaps and their Contents.-- Rock-Piles.--Indian Affection for the Graves of their Departed.--Ancient Burial-Oound on the Coast.--Rock-Walls, Embankments, and Defensive En closures.--Stone Mountain.--Fortified Towns of the Southern Indians.
turn now to the SHELL-MOUNDS. It is not an exaggeration to say that some of the islands and localities bordering upon the salt-water are hoary with these tumuli. Many are burial-mounds, while vast numbers of them are little more than the refusepiles accumulated, during the lapse of years, about the Indian settlements. Those of the latter sort--com posed of O3'ster, clam, mussel, and conch shells, the bones of deer, raccoons, buffalo, sea-turtles, large birds, and fishes, intermingled -with fragments of pottery and the debris of the encampments--remind us of those heaps to which the Danish archseologists have given the name of kjokkenmdddings. Shell-mounds formed the common graves of the Indians occu pying the coast. They abound upon all the seaislands, and are thickly congregated upon the outer bluffs and along the banks of salt-water streams. The admixture of shells imparted a permanency to many small mounds which, otherwise, would long since have been entirely obliterated. Most of them
196
AKTIQUITIE3 OF THE SOUTHERN IMDIAK3.
contain more than one skeleton; the bones being gen erally disposed in a horizontal position. In a few in stances the dead were inhumed in a sitting posture. Only occasionally do the human bones found in these tumuli indicate the action of fire.
It is well known that the Lower Creeks subsisted, to a large extent, upon oysters and fishes. Bringing oysters, conchs and clams from their natural and exhaustless beds in the adjacent creeks and marshes, they carried them to their villages and ate them. As a neces sary consequence, there occurred a rapid accumulation of shells which were carelessly thrown into heaps near the doors of their lodges. It was just as easy to use these shells in erecting mounds over the dead as to cover the skeletons with sand. That such a disposi tion was frequently made of .such refuse shells admits of no question. \Vhen we open these mounds it is not an unusual occurrence to find, intermingled with the shells and sand overlying the skeletons, the bones of large fishes, deer, and other wild animals, birds and sometimes dogs, accompanied by broken pieces of pottery, arrow-heads, flint knives, stone axes, and charred wood. The drift-shells--collected by the action of the tides into ridges so common along the coast--were also employed in the construction of these tumuli. Some are composed entirely of shells. Others are made chiefly of sand, with a layer of shells, vary ing from six inches to three feet in thickness, overly ing the whole. Others, again, appear to have been formed by the careless admixture of shells and sand, just as either material at the moment chanced to be most convenient. Others, still, consist of alternate lay ers of human bones, sand, and shells.
A sepulchral shell-mound is rarely seen above
SEPULCHRAL SHELL--MOUNDS.
19T
thirteen feet ill height. Most of them do not rise more than three or four feet alcove the plain. In form they are elliptical and circular, with base-diameters varying from ten to forty feet. As a rule, the human bones and articles deposited in them are in a better state of preservation than those found in the ordinary earthmounds on the main. The dry sand of the coast and the shell-covering afforded no mean defence against the disintegrating influences of time and the elements. So numerous are they in some localities on the sea-islands, that they mar the fertility of the cotton-fields. Multi tudes of them have been entirely levelled-by continued ploughing, and nothing but scattered shells mark the spots where they formerly stood. These tumuli afford physical proof of the general and long-continued occu pancy of the coast-region by the reel-men. A delight ful climate, frequent springs of fresh water, mild airs in winter and'cool sea-breezes in summer, fish and game in abundance, magnificent forests, and a variety of indigenous fruits, rendered this portion of the State very attractive to these improvident nomads. Appre ciating these advantages, they availed themselves of them, .and formed settlements in this section appar ently more numerous and abiding than was their cus tom elsewhere.
Tho existence of these shell-mounds is not ex clusively confined to the coast. Take, for example, that remarkable tumulus located upon Stalling's Island, in the Savannah River, more than two hundred miles from its mouth. Elliptical in shape, with a diameter, measured in the direction of its major axis, of. nearly three hundred feet, and a minor diameter of one hun dred and twenty feet, and with an average elevation of more than fifteen feet, this mound has been formed, to
19S
AKTIQUITIES OF THE SOTJTHEBN INDIANS.
a large extent, of the mussel, clam, and snail shells of this fresh-water stream. The layers of these shells are eight or ten inches in thickness, with intervening strata of sand. Human bones lie in strata. It is a huge necropolis, and contains, at a moderate calcula tion, hundreds of skeletons. It could not have been the work of a year or of a generation. It is the ac cumulation of successive and long-continued inhuma tions. There is something solemnly impressive in. the thought that "by common consent this quiet, retired, isolated, beautiful spot should have been consecrat ed exclusively to the purposes of sepulture. The absence of grave-mounds in its vicinity, the unusual dimensions of this tumulus, the numerous skeletons entombed within its bosom, all attest the fact that this mound must have been used as the general cemetery of the tribes once occupying the adjacent hills and valleys.
Removed from the noise and confusion of the vil lages, and yet so near that the bright rays of the fires nightly kindled upon either bank revealed the out lines of this island of the dead--lying not in the path trod by,the hunter--away from the conflicting voices of the council-lodge and the wild delights of the place of feasting and dancing, and yet just -where the eye of affection could ever turn and rest upon its hallo-wed form, this tumulus has stood for centuries and still stands, a convincing proof of that respect paid to their dead, and of that care bestowed upon their sepulture, which characterized the primitive peoples of these Southern forests.
"Who -will recall the associations which cluster about this silent and yet not voiceless tomb; -who enumerate the vicissitudes which have occurred since
HOUND os STAKLING'S ISLAND.
199
the first canoes, with measured dip and attendant train of mourners, landed here their precious burdens ? Whose memory "will recount the names, numbers, and deeds of those who have been here interred--who can tell the day when the first sleeper "was laid to rest, and the first shell, bright from the bosom of the Savannah, "was placed upon the new-made grave ?
The hand of the conqueror has been heavily laid upon the descendants of those who here builded this memorial of their sorrows. Even the remembrance of their former existence is fading from the recollection of those who have supplanted them in the dominion over forest, hill, and river; and yet decay--more kind than they--leaves untouched this striking monument of their affection for the dead. The forest-trees with, sturdy roots encircle this mound--their overarching branches shielding its outlines from the annihilating influences of the storm. The murmuring voices of the stream, -which so often charmed the living ear, still bring joy and gladness as in days of yore, and the song-birds still warble sweetly their morning and evening lays above these nameless dead. All else is hushed save the whispers of the wind among the for est-branches, the startled note of the solitary water fowl, frightened from its retreat among the reeds by the passing "boat, and the soothing ripple of the river. The warrior--his stout heart turned to clay, his spear heads scattered, his stone axe lying unused near his skeleton hand; the chieftain--his council-fires dead, his heroic deeds unsung, his memory forgotten; the medicine-man--his healing arts entombed, his charms crumbled into dust, his potent herbs ungathered in. the tangled brake; young man and maiden upon whose plighted troth even the cold moon beamed
200
ANTIQUITIES OF TIIE SOUTHEES INDIANS.
kindly; the care-worn mother, her toilsome journey ended--the tender infant--all rest in one common grave, and here they will remain until the last trump shall summon both civilized and savage before the judgment-seat of Him who is mightier than them all. 1
Professor Jeffries AVyman a has furnished an inter esting- account of the FEESH-WATEK SHELL-HEAPS on the St. John's Ki^er in East Florida. The distribxition of such heaps is very general. They are found upon the banks of mos-t of the fresh-water rivers of Georgia. The largest which the writer has examined are located upon the Savannah River, in Columbia County, near the confluence of the Great Kiokee Creek. Artificial in their character, they may in general terms be de scribed as the debris of the long-continued encamp ments of the natives upon the river-bluffs, while en gaged in hunting and fishing. They are frequently several hundred feet in length, and from two to five feet or more in height. Fresh-water mussels formed an important article of food with, the Indians, and were extensively gathered both for this purpose, and for the pearls which they contained. Their shells en ter largely into the composition of these heaps. In termixed 'with them are seen numerous fragments of pottery, stone axes, chisels, crushing-stones, awls, mor tars, net-sinkers, arrow and spear points, flint knives, shell beads, soapstoiie ornaments, pipes, and the bones of deer, buffalo, alligators, turtles, raccoons, of smaller animals, and of birds and fishes. Many of the larger bones are split longitudinally, as though the Indians, before discarding them, had extracted the marrow.
BEU'USE--PILES.----SHELL--HEAPS.
201
This -was done by the ancient inhabitants of Southern France, and by other primitive peoples, who, not con tent with devouring the flesh of the animals which they killed, split, or pounded the bones and sucked out the animal juices contained in them. 1
The size of these refuse-piles affords striking proof of the long-continued occupancy of these bluffs by the Indians, and their contents advise us both of the food, eaten, and the articles and implements used by these ancient peoples. Vast quantities of net-sinkers and spear and arrow points were manufactured here--the surface of the heaps being at some points covered with thousands of chips and partially-formed implements. When we come to consider the use of nets, and the different modes of fishing adopted by the Southern Indians, we "will have occasion to refer to these fresh-water shell-heaps. Refuse-piles of a kindred character have been observed all along the Atlantic coast from the Bay of Ftmdy to Cape Sable, and also upon t-he Gnlf coast. Such are extant in numbers upon the Georgia coast, indicating the favorite local ities where the Indians congregated and subsisted upon oysters, clams, conchs, fishes, and animals and birds native to the region. The particular spots occu pied by individual lodges or huts are sometimes thus perpetuated. In such instances we find the circular, depressed space formerly covered by the wigwam, sttrrorinded by a ridge or embankment of oyster-shells. These refuse-piles can be readily distinguished from the sepulchral shell-mounds.
In order to designate the grave of a remarkable warrior, who had fallen in battle, and whose body
* Sir John Lubbock's "Prehistoric Times," second edition, p. 31Y. Lon don, 18G9.
202
AKTIQTJITrES OF THE SOUTHERN INDIANS.
could not at the time be brought home by his compan ions, the Cherokees and other nations inhabiting hilly regions -were wont to cover the body of the slain with stones collected oil the spot. Every passer-by con tributed his stone to the pile, until it rose into a marked and permanent memorial of the dead. " In the woods," says Adair,1 " we often see innumerable heaps of small stones in those places, -where, accord ing to tradition, some of their distinguished people were either killed or buried, till the bones could be gathered: there they add Pelion to Ossa, still increas ing each heap, as a lasting monument and honour to them, and an incentive to great actions."
At a point where a decisive battle had been fought between the Caroliuians, under General Middleton, and the Cherokees, in which many of the latter had been killed, and the survivors compelled to abandon their settlements in the low countries and betake themselves for safety to inaccessible retreats in the mountains, Bartram' observed " vast heaps of stones," indicating the graves of the red "warriors who had perished during the conflict. Dr. Brickell * affirms the exist ence of monuments of this sort among the Carolina Indians.
In various parts of middle and Cherokee Georgia these STONE-PILES have attracted our notice. They con sist simply of fragments of rock and loose bowlders col lected from the beds of adjacent streams, or picked up on the surface of the ground, and piled one upon the other until the structure attained an altitude of from three to twelve feet. It is intimated by some of the
1 "History of the American Indians," p. 184. London, 1*775. 3 " Travels through Xorth and South Carolina, Georgia," etc., p. 346. 1792. " "Natural History ofKorth Carolina," p. 380. Dublin, 1737.
London,
MODES OF SEPULTURE.
203
early travellers that these tumuli were temporary in. their nature, and were designed merely as a protection to the bones of the dead, until they could be collected and carried home for interment iu the burial-grounds of the tribe or community of which, the deceased were members.
Within the historic period some of the NorthGeorgia tribes, imitating the custom of the Europeans, dug graves in the earth three or four feet deep, lining the bottom and sides with poles and bark. The corpse, enveloped in a blanket, was then carefully laid in this rude coffin, a cover of bark and poles being placed above, so as to protect it from contact with the restored earth. A-ffcer the grave was filled, stones were added to give shape and permanency to the place of sepulture. lThe custom of depositing "with the dead all articles of use and ornament was scrupulously observed. Various are the articles of European manufacture which have been obtained from these later graves.
Frequently the body -was hiddeu away iu some fis sure of the rocks, or in the hollow of a tree--the en trance, in each instance, being securely closed.
They often interred beneath the fioor of the cabin, and then burnt the hut of the deceased over his head, consuming such personal property as was not lodged in the grave, and thus obliterating all traces of the inhumation.
A-t other times, apparently to avoid the trouble of sepulture, the dead bodies were thrown into some neighboring river.'
Intercourse "with swindling European traders caused the Indians to neglect those laborious rites of sepul ture which at an earlier period "were religiously ob-
1 "Memoirs of Lieutenant Timberlake," p. 67. London, 1765.
204
ANTIQUITIES OF THE SOUTHERN" INDIANS.
served. In this respect, as in almost every other, they became indifferent and demoralized ; and yet up to the period of their removal from the State, they cherished an abiding attachment for the graves of their kindred and chiefs. The idea of abandoning them was per haps the most difficult they could be induced- practi cally to entertain. " Why," asks the Viscount de Chateaubriand, 1 " are the savages of America, among all the nations of the earth, those who pay the greatest veneration for the dead? In national calamities the first thing they think of is to save the treasures of the tomb ; they recognize no legal property but where the remains of ancestors have been interred. "When the Indians have pleaded their right of possession, they have always employed this argument, which, in their opinion, was irrefragable : ' Shall -we say to the bones of our fathers--Kise and follow us to a strange land ?' Finding that this argument was disregarded, what course did they pursue ? They carried along -with them the bones which could not follow.
" The motives of this attachment to sacred relies may easily be discovered. Civilized nations have monu ments of literature and the arts for memorials of their country; they have cities, palaces, towers, columns, obelisks; they have the furrows of the plough in the fields cultivated by them; their names are en graven in brass and marble; their actions are re corded in their chronicles.
" The savages have none of these things; their names are not inscribed on the trees of their forests; their huts, built in a few hours, perish in a few mo ments ; the -wooden spade with, which they till the soil has but just skimmed its surface without being
1 " Travels in America and Italy," vol. i., p. 215. London, 182S.
BUBIA1--GKOUND O1T THE GEORGIA COAST.
205
capable of turning up a furrow; their traditional songs are vanishing with the last memory which re tains, with the last voice which repeats them. For the tribes of the New World there is, therefore, but a single monument--the grave. Take from the savages the bones of their fathers, and you take from them their history, their laws, and their very gods; you rob these people in futtire times of the proof of their exist ence, and of that of their nothingness."
But a short time since "we stood in the midst of an ancient and extensive Indian burial-ground on one of the low-lying islands which fringe the Georgia coast. Earth and shell mounds were thickly congregated on every hand. A bold spring issuing from a sandy bluff--adjacent salt-water streams and wide-spread marshes filled -with oysters, crabs, and fishes, and neighboring forests once abounding with game--ren dered this, in the olden time, a spot highly attractive to the reel-men. The solemnity of death and of deso lation--so far at least as this entombed race was con cerned--rested upon every thing. Even the tradi tions of the locality were forgotten, and the grand old live-oaks -which knew these sleepers during their "wak ing hours whispered no legends of their customs, their wars, their loves, their lives, or their deaths. Theii' feeble " footprints on the sands of time" had been obliterated, by the tread of a statelier civilization, and there were none to care for their graves. The same sun was sinking to his rest. The breath of the myrtle and the orange still perfumed the ambient air. Kin dred waves -washed the bermuda-covered shore and dashed their spray, as in former days, against the roots of the vine-clad cedars. Eagles of the same bold flight soared majestically in the tranquil heavens, and contig-
206
ANTIQUITIES OF THE SOUTHEEH INDIANS.
nous woods were vocal with the notes of birds native liere for centuries. The same "blue sky, the same soft sea-breezes, the same generous mother earth, kindred forests and ilowers, the same loves and voices of Nature, but all else how changed ! The living Indian frequented no more his favorite groves. Autumnal leaves long ago covered the last trace of his rude hut. His watch-fires were dead. His council-lodge years ago mouldered into utter decay. His village was con verted into a cotton-field, and the ploughboy tram pled "upon and furrowed mound-tombs hallowed by unrecorded memories of chiefs, -warriors, priests, medi cine-men, and the nameless dead of tribe and family. Never more -will weeping mother -with trembling hand fashion the funeral-vase. The sorrowing circle will never again assemble around the sepulchral fires, nor stalwart arms above the ashes of the dead heap the grave-mound. Beaten upon by the rains and wasted by the winds, there "will soon be scarce a vestige of these tumuli. Few, if any, -will gather up and deposit in some secure resting-place these neglected bones as they whiten in the sun and crumble into dust amid the fields of the present owners of the soil.
" Mors sola fatetur Quantula sirit kominum corpuscula."
The world, -waxing old, forgets the names, palaces, pyramids, and sky-searching towers even of those who once held mighty sway over vast domains; and, in the wreck of ages whole nations, living and dying without letters, are remedilessly engulfed in the great ocean of oblivion.
As -we mused amid these silent, storm-beaten graves, the mournful strains of the Coplas of Man-
ROCK--"WALLS, ENCLOSURES, ETC.
207
rique entered witli peculiar pathos into our saddened tli oughts.
" Our lives are rivers gliding free To that unfathomed, boundless sea-- The silent grave. Thither all earthly pomp and boast Koll, to be swallowed up and lost In one dark wave. Thither the mighty torrents stray, Thither the brook pursues its way, And tinkling rill. There all are equal. Side by side, The poor man and the son of pride Lie calm and still."
^fife conclude this account of the more prominent traces of early constructive skill "by an allusion to the existence of ROCK-WALLS, EMBANKMENTS OF EARTH, AND ENCLOSURES which were designed, we think, principally for defensive purposes. The circnnivallation by means of which the top of t: Brown's mouud" was fortified has already been mentioned and described.
About half-way up Stone Mountain, in De Kalb County, -where the acclivity becomes very marked as one ascends the western slope, on both sides of the usual pathway are the remains of a rock-wall which was originally intended for the protection of the upper portion of the mountain. This wall is still in some places two feet high, and is composed of fragments of rock, all capable of manual amotion, piled one upon the other. At either end this wall extended to the precipitous sides of the mountain where--its defensive presence being no longer necessary--access to the sum mit was either altogether denied or rendered so difntcult and perilous as to preclude the possibility of any thing like a combined attack. Where the approach to
208
ANTIQUITIES OJ? TUB SOUTHERN INDIANS.
the tipper pa,rt of the mountain was most facile, and where, by common consent, a path or trail seems to have been established, an opening occurs in the wall. This opening or gate-way wa.s effectually commanded by a ledge of rocks a little higher up the mountain and directly in front of the gap, constituting a formidable natural breastwork from which, in all security, the defenders could have launched their arrows and spears against an enemy seeking to force a passage along this, the most practicable route up the mountain. In anti cipation of an attack, this opening could have been rapidly closed, and thus the entire defensive line effect ually established. Below this circurnvallation are numerous fragments of rock which, originally forming a part of the wall, have, in the course of time, become detached and entirely separated from it.
This huge pile of granite, towering in naked grand eur far above the adjacent valleys, was a favorite resort of the Indians during the summer months. In many places upon the summit of Stone Mountain may still be seen the indications of this former occupancy.
Similar rock-walls exist upon Mount Yonah, and guard the summits of other solitary peaks within the confines of Georgia. From their number and location it would appear that these fortified mountain-tops con stituted the retreats of the natives when sore pressed in the plains. Protracted sieges were then unknown, and in the nature of things impossible. Hence, pur suit was speedily abandoned -when the advance was interrupted by formidable barriers of this descrip tion.
jSTor were these rock-defences confined exclusively to the mountains. They sometimes appear in the
EAKTD--EMBANKMENTS.-----FOKTIITED TOWNS.
209
valleys, and are circular, quadrangular, or irregular in shape according to the physical conformation of the locality for the protection of which they were erected.
"We note also embankments of earth from two to four feet high and from three to five feet in width,'generally circular in form and sometimes semilunar in shape--in the latter case the horns extending to and resting upon some stream. "Within such enclosures are embraced areas varying in size from two to twenty acres, and it is suggested that, in many instances, these parapets formed the foundations in which were securely embedded the lower ends of the stockades \vith which the Southern Indians were wont to fortify their prin cipal towns. Less nomadic in their habits than the Northern and "Western tribes, and bestowing no little attention upon the cultivation of maize, the southern nations rendered permanent their seats and protected their homes against the incursions of wandering bands who from time to time sought to dispossess them of their cleared fields, their fish-preserves, and their substantial granaries,
In plate xxx. of the Brevis Narratio,' De Bry furnishes its with a spirited sketch of a Availed town built by the Florida Indians. The following is a translation of the accompanying text : " The Indians build their towns in this wise. Having made choice of a spot near a running stream, they level it off as evenly as they can. They next draw a furrow of the size of the intended town, in the form of a circle, in which they plant large round stakes, twice the height of a man, and set closely together. At the place where the entrance is to be, the circle is somewhat drawn in
1 Francoforti ad Mcenum, anno 1591.
210
ANTIQUITIES OF TUB SOUTHEBN INDIANS.
after the fashion of a snail-shell, making the opening so narrow as not to admit more than two at a time. The bed of the stream is also turned into this entrance. At the head of the entrance a small round building is usually erected: within the passage is placed another. Each of them is pierced with slits and holes for observation, and is handsomely finished off after the manner of the country. In these guard-houses are placed those sentinels who can scent the trails of ene mies at a great distance. As soon as their sense of smell ing tells them that some are near, they hasten out, and, having found them, raise an alarm. The inhab itants, on hearing the shouting, immediately fly to the defense of the town, armed with bows, arrows, and clubs.
" In the middle of the town stands the king's palace, sunlc somewhat below the level of the ground, on ac count of the heat of the sun. Around it are ranged the houses of the nobles, all slightly covered with palm * branches; for they mate use of them only during nine months of the year, passing, as we have said, the other three months in the -woods. When they return, they take to their houses again; unless, indeed, they have been burned down in the mean time by their ene mies, in which case they build themselves new ones of similar materials. Such is the magnificence of Indian
palaces." In plate xix. of the " Admiranda ^Narratio," we
have a plan of the town of Pomeiooc, and are informed that while the villages of the Virginia Indians were also defended by stockades, the poles inserted in the ground were smaller and less strong than those used by the Florida tribes. Both the Gentleman of Elvas and
FORTIFIED TOWNS.
211
Hernandez de Biedma allude to the existence of stock aded forts defended by the natives. 1
Du Pratz," speaking of the Louisiana Indians, says: " Wlien a nation, is too "weak to defend itself in the field, they endeavor to protect themselves by a fort. This fort is built circularly of two rows of large logs of wood, the logs of the inner row being opposite to the joining of the logs of the outer row. These logs are about fifteen feet long, five feet of which are sunk in the ground. The outer logs are about two feet thick, and the inner about half as much. At every forty paces along the wall a circular tower jets out; and at the entrance of the fort, which is always next to the river, the two ends of the "wall pass beyond each other 'and leave a side opening. In the middle of the fort Stands a tree with its branches lopt off within six or eight inches of the trunk, and this serves for a "watch tower. Hound this tree are some huts for the protec tion of the women and children from random arrows; but notwithstanding all these precautions for defence, if the besieged are but hindered from coming out to water, they are soon obliged to surrender."
The town of Manilla, where De Soto's army en countered such determined resistance and loss at the hands of the Alibamons, was strongly fortified by piles driven in the ground " with timbers athwart, rammed with long straw and earth between the hollow spaces," so that the work, in the language of Herrera, " looked like a wall smoothed with a trowel." At intervals of eighty paces were towers in which eight men could
1 " Narratives of the Career of Hernando de goto," translated by Buckingham Smith. No. V. Bradford Club Series, pp. 99, 100, 248. New York, 1866.
a " Ilistory of Louisiana," etc., vol. ii., p. 251, etseq. London, 1763.
2L2
ANTIQUITIES OF THE SOUTHERN INDIANS.
fight. These were loop-holed, and. the town was en tered by means of two gates.
It is probable that most of the earth-walls or parapets in the valleys, with traces of an exterior ditch and sometimes of an interior trench, also, indi cate simply the locations of the palisades planted for the protection of ancient towns. The upright position and defensive power of these posts, inserted in the ground, would have been materially strengthened by a bank of earth thrown up on both sides of the stock ade ; and nothing would, be more natural than the presence of ditches or trenches, both within and with out, whence material was obtained for this purpose. In these enclosures the position of the gate-ways is often quite distinct.
CHAPTER X
Implements and the Jane-Matting.--Shell
IN the upper part of Nacoochee Valley, and near its western extremity, is a prominent earth-mound. Located not far from the Chattahoochee River, and ris ing some twenty feet or more above the surface of the surrounding valley, it has long constituted a marked feature in this beautiful region. For many years its slopes and summit have been cultivated, and, within the recollection, of the older inhabitants, this tumulus has lost much of its original dimen sions. Elliptical in shape, it has a flat top, declin ing somewhat toward the southwest. Measured in a northeasterly and southwesterly direction, at right angles, its base-diameters are, respectively, one hun dred and ninety, and one hundred and fifty feet; -while its apex-diameters, ascertained in the same direc tions, do not fall short of ninety and sixty feet. It is entirely artificial, and appears to be -wholly composed of the earth gathered from the neighborhood of its base. There are no terraces, the sides sloping gradxially from the summit. Tradition has preserved no
214
ANTIQUITIES OF THE SOUTHERN INDIANS.
memories of the people by whom it was erected, and its treasures, if any, are still concealed within its own
bosom. In June, 1870, Oapt. J. II. Nichols, while ploughing
in the vicinity of this tumulus, discovered, several inches below the surface of the field, a number of large stone slabs. They were lying at a remove of about thirty feet from the -western slope of the mound. At a loss to account satisfactorily for their presence ill this locality, and his curiosity being ex cited, he set about removing them. During the prog ress of the investigation, he unearthed three stone graves, quite near each other, but not disposed in a uniform direction. These graves were parallelogrammic in shape, being, seven feet long, three feet "wide, and a little more than two feet and a half deep. They were all filled "with earth, and the surface of the field above them was somewhat elevated beyond the level of the surrounding valley. The sides consisted of rough slabs of slate, between two and three feet long, and about two feet wide, set up on end. The bottom of the central grave was paved "with oval bowlders which had evidently been obtained from the bed of the Chattahoochee. But one of the three--and that the central grave--was covered. For the covering, or lid, flat slabs of stone rather more than three feet in length had been employed ; so that when they rested upon the upright sides and ends of the grave, the en closure of this vault or rude sarcophagus was com
plete. In this central grave a male skeleton, measuring
more than six feet, lay extended at full length. Each of the other two graves contained the bones of more than one skeleton lying in disorder, and carelessly
STONE .GEAVES IM" KACOOCIIEE VALLEY.
215
piled in without any regard to regularity. It was obvious that these bones were in a detached condition when they were placed in these enclosures. It seemed impossible from them to construct distinct and com plete skeletons. When removed from the graves and exposed to the air, most of them crumbled. Further investigation will probably develop the existence of other stone graves of similar construction in this vi cinity.
So far as we are informed, these are the first an cient stone graves which have been observed within the geographical limits of Greorgia. "We have already seen that shell and earth mounds abound along the coast. The valleys of the Savannah, the Chattahoochee, the Etowah, the Oostenaula, the Alatamaha, and of other rivers, are rendered remarkable by the presence of tumuli august in their proportions. Even the lonely pine-barren region is not wholly wanting in these proofs of the former occupancy of the red race. In Cherokee Georgia heaps of stones desig nate the last resting-places of the Indians, while a cleft in the rock, a hollow tree, or a small moiincl often formed a hiding-place for the dead. In other portions of the State regular inhumations occurred with but slight external marks to commemorate the places of sepulture. Although it was confidently believed that the stone-grave makers of the Tennes see and Cumberland Valleys might have crossed the mountains which intervened and possessed themselves of the pleasant valleys of Georgia, the fact that they had actually done so, and, in accordance with their es tablished custom, deposited their dead in rude sar cophagi in these localities, was never fully established until by the recent investigations of Capt. Nichols.
210
ANTIQUITIES OF THE SOUTHERN INDIANS,
Bartram ' observed in the environs of Keowe, " on the leases of the rocky hills immediately ascending from the low grounds near the river bank, a great number of very singular antiquities, the work of the ancients." They were between four and five feet in length, two feet high, three feet wide, and con structed of four flat stones--two set on edge forming the sides, a third closing one end, and a large flat stone placed horizontally on the top of tliese three completing the enclosure--the other end being left open. He could not determine whether they were ovens, sacrificial altars, or sepulchres, and all inquiries failed to elicit any definite information with regard to their \ises, both from the Indians and the trader who accompanied him. These structures "were upon the surface of the ground and varied in their dimensions.
To Mr. Haywood' we are indebted for early no tices of the existence of stone graves in Tennessee, at not a great remove from the boundary-line of (reorgia. Professor Troost found them in the Cumber land Valley, and described them " as rude fabrics com posed of rough flat stones (mostly a kind of slaty lime stone, or slaty sandstone, both abundant in our State). Such flat stone was laid on the ground in an excava tion made for the purpose; upon it were put (edge wise) two similar stones of about the same length, as the former, and two small ones were put at both extremities so as to form an oblong cavity, lined with stones, of the size of a man; the place for the head and feet had the same dimensions. When a coflin was to be constructed next to it, one of the side-stones
1 "Travels," etc., p. 370. London, 1792. " " Natural and Aboriginal History of Tennessee," pp. 123, 201-20Y. Nash ville, 1823.
STONE GRAVES.
217
served for Tbotli, and consequently they lay in straight rows, in one layer only. I never found one above the other." 1
On the banks of the Merameg, about fifteen miles above the confluence of that river with the Mississippi. Mr. Say in 1819 observed numerous stone graves. They did not rise above, the general surface of the ground, but their presence was readily ascertained by the projecting vertical stones which enclosed them. The sides of the graves were neatly constructed of long flat stones vertically implanted and adapted to each other, edge to edge, so as to form continuous walls. Their coverings consisted of flat stones placed horizontally above them. These graves varied in length from three to six feet. They were filled "with earth, and the bones which they contained appeared to have been deposited after they had been separated from the flesh, and from each other, in accordance with a custom 'which obtains among some Indian tribes even to the present day. In some of these vaults rude pottery was found. It had been represented that these graves contained the skeletons of a diminu tive race of men, but a careful examination of their contents proved conclusively the utter falsity of the statement. ]STear the city of St. Ix>nis more graves of this description were observed. Mr. Say express es the opinion that these sepulchral chambers are more modern than the tumuli which abound in this region.1
In the State of Missouri, between the river A_nx Vaix and the Saline, on the farm of a Mr. Bogy, are
1 Transactions American Ethnological Society, vol. i., p. 359. a "Account of an Expedition from Pittsburgh to the Rocky Mountains," etc., compiled from the Notes of Major Long, etc., vol. i., pp. 55-57. London, 1823,
218
ANTIQUITIES OF THE SOUTHERN INDIANS.
many small elevations, evidently artificial, with, trees from fourteen to eighteen inches in diameter growing upon them. They contain graves, the outlines of which are formed of sharp stones standing on edge and sloping inward at the bottom. 1
In May, 1843, Dr. A. -Wislizenus visited and ex amined quite a number of stone graves in the neigh borhood of Prairie du Rocher, Randolph County, Illinois, three miles east of the Mississippi River, and not far from old Fort Chartres.
His description is as follows: in general construc tion they were coffin like--their side-walls, top, and bottom, being formed by nat limestones joined to gether without cement. The size of the grave was adapted to that of the person to be buried in it, vary ing in length from one and a half to seven feet, in width from one foot to eighteen inches, and in depth from one foot to a foot and a half. The top layer of stones was seldom deeper than half a foot below the ground. Although located near each other, no order was observed in the position and direction of these graves."
To Professor Charles Rau I am indebted for the following memoranda of his researches among the stone graves of Illinois :
Indian cemeteries are of frequent occurrence in the " American Bottom," which extends along the bank of the Mississippi, in the State of Illinois, and is bounded toward the east by the picturesque " Bluffs "--an ex tended range of elevations indicating the former left bank of the " Father of Waters." These cemeteries are usually found on the brow of a hill, in accordance
1 See Bulletin Araerican Ethnological Society, vol. i., pp. 49, 50. 3 See Transactions St. Louis Academy of Science, vol. i., pp. 66, 67.
STONE GRAVES IN ILLIWOIB.
219
with the custom of tlie Indians to select elevated places for jurying their dead. The graves consist of rough limestone slats, loosely put together at right angles, so as to form a kind of stone coffin enclosing the corpse on all sides. The bottom, sides, and cover, are all formed of stone slabs. Rectangular in shape, these graves vary in length according to the size of the occupant. Their average depth is about three feet--the top stones being covered -with earth. The side-slabs protruding a few inches above the ground indicate the single graves. These are often arranged in rows con tiguous to each other, but are sometimes distributed without any view to regularity. You may see, for instance, at the same burial-place, six or seven, graves in a row, and a few others joining them in a quite unsymmetrical way. No fixed rule prevailed in the location of these graves with reference to the cardinal points. Professor Rau examined a group of seven or eight situated on a high eminence of the Bluff, a mile northeast of the conical rock formation known as the " Sugar Loaf," near the town of Columbia, in Monroe County, Illinois. One of these graves was nearly quadrangular in shape, measuring between five and six feet each way. .After removing the covering he found a skeleton in a rather decayed state, lying flat on the bottom stones, with the arms not extend ed along the side of the body, but stretched out at right angles with it. Hence the unusual width of the grave. The skeleton was that of a medium-sized individual; skull not very large, and the teeth, al though much worn, in excellent condition. 'Not only the grave but also the skull, and even the hollows of the bones were filled -with earth. No trace of any manufacture appeared in this grave.
220 . ANTIQUITIES OF THE SOUTHERN INDIANS.
In anotlier cemetery similar to that just alluded to, but containing more graves, situated on the spur of the Bluffs, five miles south of Columbia, decayed human bones were seen embedded, without order in various portions of the earth which filled one of the graves. Fragments of a skull were found quite dis tant from each other. From the thinness of the frag ments and their curvature, it was conjectured that the remains were those of a child perhaps ten or twelve years old. Nothing of artificial origin was discovered in this grave. Obviously it had never been disturbed, for the ground within the stone enclosure was very harcl and traversed by roots as large as a man's arm. In this instance the bones must have been interred after the flesh had been removed. It is a fact well remembered by many persons in this neighborhood that the Indians who inhabited this region during the early part of the present century (probably Kickapoos) buried their dead in stone coffins. Dr. Shoemaker, "who resided on a farm near the town of Columbia, in 1861, showed Professor Rau, in one of his fields, the empty stone grave of an Indian "who had been killed by one of his own tribe, and there interred "within the recollection of some of the old farmers of Monroe County. The skeleton, which had been exhumed a few years previously, "was in a good state of preserva tion, and Dr. Shoemaker used the skull in imparting practical instruction to a medical student then in his office.
Other small cemeteries of a like character were ob served by Professor Ran in the neighborhood of " Sul phur Springs," in Jefferson County, Missouri. In them several food-vases "were found.
By far the most extensive investigations, however,
STONE GEAVES IH" TENNESSEE.
221
are tliose recently conducted by Professor Joseph Jones, J\I. D., in the Cumberland Valley and in various other localities in the State of Tennessee.
Most of the stone graves, examined by Dr. Jones, were parallelograminic in form. Some of them "were coffin-shaped, others were square; and, in one instance, in the centre of a mound, he observed, a hexagonal stone grave, with parallelogrammic stone graves radi ating on all sides from it. Some graves were only ten inches long, and five inches wide. These contained the bones of infants. The largest he saw were about eight feet long-, and two feet and a half wide. Inter mediately, were graves of all sizes. Their depths va ried from teu inches to a foot and a half. As a rule the larger graves contained but a single skeleton. In the square graves it was not unusual to find portions of more than one skeleton: for example, two skulls were not infrequent. The flesh had evidently been removed from the bones before they "were placed in these receptacles. On more thau one occasion he no ticed the bones of the toe inserted in the nasal open ings of the skull. The body was never enclosed in a sitting posture. The square graves appeared to be the common receptacles for the collected bones of the dead. In the centre of the sarcophagus the skull was often located, and the long bones of the skeleton were ar ranged around it. In many instances no order was observed in the collocation of the bones. Dr. Jones saw acres of stone graves in several of the valleys of Tennessee.
ISTear Brent-wood, twelve miles from Nashville, he opened a mound which was composed entirely of stone graves, located one above the other, to the height of four tiers in the centre. Those lowest in order -were
2:22
ANTIQUITIES OE THE SOUTHERN INDIANS.
uniformly square graves. Just above them were long and square graves, while the two upper tiers consisted of graves between six and seven feet in length, and about a foot and a half wide. The top stones of the highest graves 'were so arranged as to form a continu ous stony covering for the entire mound. This tumu lus, which was but an aggregation of individual graves, was seventy-five feet in diameter, and between six and seven feet high. It was covered with a layer of earth, several inches thick. This unique mound was located near the centre of an enclosure, some ton acres in ex tent, surrounded by an earth-wall, three feet high, at the date of his visit.
The valleys of the Cumberland, the Harpeth, Duck, and Stone ^Rivers, teem with the sepulchres and monu ments of the stone-grave makers.
From these graves Dr. Jones obtained numerous stone and clay images, marine shells, stone imple ments, arrow and spear heads, a stone sword, agricul tural implements, various ornaments of stone, clay, and shell, pots, vases of curious devices, and copper crosses. This collection possesses rare value for the student of American archaaology. It is very rich in crania.
Upon some of the bones taken from these graves the ravages of syphilis were unmistakable. Prom various indications which were satisfactory to his own mind, Dr. Jones "was convinced that inhumations had occurred in these rude vaults since the period of primal contact between the Europeans and the red race.
It is perhaps not unlikely that the Chaouanons con structed many of these Tennessee graves, and, crossing the mountains which intervened, peopled Nacoochee
STONE GRAVES.
223
Valley and other portions of Georgia. Tlie Shawnees, Shawanoes, TJtchees, and. Sauvanogees or Savan nahs, at some remote period may have acknowledged allegiance to this race--a people from which sprang some of the noblest specimens of the red-men of whom we have any knowledge. The TJtchees claimed to be autochthons, and always contended that they were the original proprietors of the soil. It is not too much to expect that future investigations will confirm the conjecture that stone graves will be found in the valleys of the Chattahoochee, the Etowah, the Oostenaula, the Coosa, and perhaps the Sa vannah.
The custom of reserving the bodies until they had accumulated sufficiently to warrant something like a general inhumation, and the practice of turning over the corpses to certain persons, who answered in a rude "way to the calling of undertakers, that they might strip the flesh from the bones and enclose the latter in bark coffins until the set time of burial occurred, obtained, as we have already intimated, among more than one of the j^orth _A-merican tribes.1
An examination of the stone graves of !N"acoochee Valley inclines us to the belief that to the prevalence of some such custom as this are the two graves in debted for the remains of several dead enclosed within them. The lack of order in the disposition of the bones, and the careless commingling of various por tions of several skeletons, are evident, while in the central grave the corpse was carefully laid at full length upon the stone flooring. As "we proceed we will perceive additional reasons for conjecturing that
'"Travels of William Bartram," p. 514. London, 1702. "Travels of J. Carver," p. 402. London, 1778.
224
ANTIQUITIES OF THE SOUTHUBN INDIANS.
tliis grave formed the receptacle of some cliief or warrior of note. He -was an old man of great stat ure. The few teeth remaining in the lower jaw were much "worn, and the alveolar processes had been greatly absorbed. Unfortunately, the skull -was in such a decayed condition that it could not be preserved.
It is a fact worthy of note that stone graves, not unlike those "which we have been considering, have been found in England, Scotland, Germany, France, and in other portions of the world. They are the simplest forms of tombs; and, because of their dura bility and the facility with which they could be con structed, very naturally commended themselves to the use of such as were jealous of the bones of their dead. In a word, they may be described as sepulchral cham bers or stone chests--either rectangular or approach ing polygonal forms, in consequence of the rough and misshapen character of the materials employed--roofed with blocks of Nature's own hewing.
As we have remarked, each of these graves con tained human remains. In the central grave was the skeleton of an old man more than six feet high. This corpse had been carefully deposited upon the floor of the vault, at full length, the arms lying paralled with the body. In the other two graves the bones had been disposed -without any regard to regularity. Por tions of several skeletons were found in each, and it was evident that they had been inhumed in utter dis regard of every thing savoring of order. None of these graves had been disturbed previous to this ex amination. Although located in a cleared field, which had been cultivated for a number of years, the plough share had never before touched the stone covering which sheltered them.
COPPEK IMPI>:KMENT.----CAjSTE-MATTING.
225
The most remarkable object found in the central or chieftain grave, was a OOPPEB TMPT/EMENT. It lay near the shoulder of the skeleton, and beneath it was a piece of cane-matting, probably the remnant of the sheath or basket which enclosed it when first deposited in the grave of its owner. The only portion of the matting or basket-work in condition to be removed and preserved was that part which was immediately underneath and in contact with the implement. It was discolored by the oxide of copper, which exerted a conservative influence. This sheath or matting 1 consisted of thin layers of split cane, about the quar ter of an inch in width, interwoven at right angles with each other. The cane had been prepared for the purpose, by being split into strips of uniform width. From these the softer, inner portions had been removed, so that only the thin, hard, outer sur face remained. Those at all familiar with this reed will readily remember how very firm and almost indestructible by ordinary exposure its tough integnment is. The use of this material by the Cherokees in the manufacture of baskets and other articles of orna ment and domestic value, was continued until a late period. Adair says: "They make the handsomest clothes-baskets I ever saw, considering their materials. They divide large swamp-canes into long, thin, narrow splinters, which they dye of several colours, and manage the workmanship so well, that both the inside and out side are covered with a beautiful variety of pleasing figures." " Traces still exist, indicating that the strips of cane composing the piece of matting we are now
1 See Fig. I, Plate VI. 2 lt History of the American Indians,11 p. 424. London, 1775.
idb
ANTIQUITIES OF THE SOUTHERN INDIANS.
considering were originally dyed, some of them black, and others yellow. The moisture of the earth, the de posit of the oxide of copper, and the gnawing tooth of Time have sadly interfered with the primal coloring. This is probably nothing more than the fragment of a cane-basket--then in general use among the Indians --which was deposited in the grave at the time of the inhumation. Possibly the copper axe and other arti cles of the deceased were then placed in it.
The copper implement, which is an object of un usual interest, is nearly ten inches in length, two inches and three quarters wide at the cutting edge, and two inches wide at the upper or helve end. The cutting edge is arching, while the other end, except at the corners, is square. It possesses an almost unifoi'in thickness of a little less than the tenth of an inch, and "weighs nine and three-quarter ounces avoir dupois.
An inch and a quarter from the upper end, and ex tending diagonally across the implement, is a smooth, "worn space on each side, about an inch and a quarter in "width, showing where and how this axe was in serted in its handle. "W^e can determine the precise angle of inclination "which the axe sustained to the han dle. The abrasion caused by the handle is very dis tinct. This implement is made of pure copper, and the lamination is clearly discernible. That it had been used, is evidenced both by the abrasion caused by the handle, and also by the fact that the cutting edge is some what split and broken--the implement being other wise perfect. So thin, however, is this axe, it seems scarcely provable that it could have been applied to
1 See Fig. 2, Plate VI.
COPPEB IMPLEMENT FKOM STONE GKAVE.
227
any general practical uses. The material of which it is made being pure, uative copper, if subjected to violent contact with any Lard substance, would ne cessarily bend and prove comparatively valueless. We think it was carried as a badge of distinction and treasured as a valuable ornament or possession, and not employed as a weapon of war or used for incisive purposes. Manufactured of native copper, it ^was beaten, into its present form without the in tervention of heat. In its construction the workman regarded his material as a sort of malleable stone, dealing -with it as such, and not as a metal capable, under the influences of heat, of being readily hammered into the desired shape. The surface of this axe is con siderably oxidated, except wliere it was surrounded by the handle, which -would indicate not only that it was attached to the handle at the time of its inhumation, but also that the handle must have consisted of some hard substance, which lasted for a long period subse quent to the inhumation and thus protected the in serted portion of the implement from those influences which operated to oxidate the exposed surface. The handle had worn that portion of the axe which it en closed quite smooth; and this fact, while evincing no inconsiderable use, tended to render such part least liable to decomposition or oxidation. No trace of the handle remained in the grave. Clavigero says the Mexicans had copper axes, -with which they cut trees, and that they were inserted in an eye of the handle. In a similar way was this axe attached; lashings of deer-sinews, bark, or buck-skin being used to keep it securely fastened. 1
1 See " Ancient Monuments of the Mississippi Valley," p. 198, Fig. 83.
228
ANTIQUITIES OF THE SOUTHERN INDIANS.
It will be remarked that the peculiar features of this Nacoochee axe are its length, it unusual thinness, and the existence of a clearly-defined space on each side, showing not only that it "was inserted in the eye or split of a handle, but also the precise point and angle at "which it "was so enclosed and held in position. Compared with the Chillicothe axe,' the Long-Island axe, and others "which might be mentioned, it will readily be perceived how materially the present axe differs from them all both in shape and weight. So far as our knowledge extends, this specimen is unique in more than one particular.'
Copper implements are rarely found in Georgia. The present is the finest specimen, which, after no mean search, has re-warded our investigations. Native copper exists in portions of Cherokee Georgia, Tennes see, North Carolina, and Alabama, but it is generally found in combination with sulphur, and not in a mal leable form. We are not aware of any locality, among those enumerated, "whence the Indians could, have secured that metal in either quantity or purity suffi cient to have enabled them to have manufactured this
implement. If we may credit the accounts of the early voyagers
and adventurers, the tribes of this region, at the times when the Europeans first visited them, "were possessed
of but little copper. Sir "Walter Raleigh's companions observed copper
ornaments in the hands of some of the Indians of the coast. Of the many mounds, however, which the writer has carefully opened and examined along the
USE OF COPPER BY THE SOUTHERN INDIANS. 229
coasts of South Carolina, Georgia, and Florida, not one contained a single copper implement or ornament of native manufacture.
In the narrative of the first attempt of the French, under Captain John Ribault, in 1562, to colonize the newly-discovered country of Florida, mention is made of the fact that the natives spoke of mines of cop per in the mountains of Appalatcy.1 It has never, we believe, l>een satisfactorily ascertained whether this reference was to particles of gold or copper; and the stream of gold " said "to be issuing from the foot of the mountains had in all likelihood no surer phys ical existence than the fountain of perpetual youth, conjectured, longed for, eagerly sought, but undiscov\ered amid the everglades of the " Land of Flow ers." The Fidalgo of Elvas alludes to the circumstance that the Indians informed the Governor De Soto of the existence, at Chisca, of a foundery of gold and cop per, but makes specific mention of no copper imple ments in the possession of the natives, except some chopping-knives at Cutifachiqui, -which were thought to have a mixture of gold in them. 3
Cabega de Vaca says : " Among the articles that were given to us, Andres Dorantes received a bell of copper, thick and large, figured with a face, which they (the Indians) had shown, greatly prizing it. They told him that they had gotten it from others, their neighbors ; and we asking them -whence they had obtained it, they said that it had been brought from the direction, of the north, where there was much cop per, and that it was highly esteemed. We concluded
230
ANTIQUITIES OF THE SOUTIIEEK INDIANS.
that whencesoever it came, tliere was a foundry and that wort was done in hollow form."
A-t a subsequent period, upon showing this bell to other Indians, their response -was, " in the place whence that had come there were many plates of the same material, and. that it was a thing they greatly esteemed." '
But one other allusion is made to copper, in this narrative, and it is this ; when in the prairie country of Texas, Cabeca de Vaca saw a copper article, which had been fashioned by the' natives in the form of a " hawk-bill."
In the Portuguese narrative the Indians are said to have obtained pearls from the beds of the interior riv ers, which they pierced "with heated copper spindles and strung around their necks, arms, and ankles.
A-t each of the three gates of the Temple of Talomeco, three miles distant from the town of Cutifachiqui, were stationed gigantic wooden statues, variously armed with clubs, maces, canoe-paddles, copper liatcliets, drawn bows, and long pikes. These implements were ornamented with rings of pearls and l>ands of copper.'
On the bank of the Mississippi Kiver, Hennepin was courteously received by an Indian chief, clothed in a " kind of white gown," which women had spun of the bark of trees. Before him two male attend ants carried a " tlvirt, plate of copper as shining as gold."
Harlot, Captain John Smith, and others, allude
1 " Narrative of Alvar Nunez Cabeca de Vaca," translated by Buckingham Smith, p. 92. Washington, 1851.
2 G-arcilasso de la Vega, pp. 274, 282. 3 Hennepin's "New Discovery," etc., p. 156. London, 1698.
COPPER REGARDED AS A MAI/LEABLE STONE. 231
to the presence of " diuerse small plates of copper," in the possession of the Virginia Indians; and De Bry has figured ornaments of the same metal worn by the natives of Florida.
Without multiplying these references, it may, we think, Ibe confidently asserted that while the early writers note the presence of copper implements and ornaments among the Indians of this region, their nar ratives prove that such were comparatively rare and highly prized by the natives. It is not shown that they were manufactured here, and we very much doubt whether there was in the Southern States (or within the geographical limits at present embraced by them) a natural vein or deposit of copper, accessible to the Indians, of sufficient size and purity to have afforded them the material necessary for the fashion ing of such an implement as that now before us. Our impression is, that the metal of which the Nacoochee axe is formed was obtained from the shores of Lake Superior, and that probably the implement itself was there made.
The art of melting copper was neither understood nor practised by the natives. Of the method of work ing it "while in a heated state, the primitive artist ap peared to be entirely ignorant. Regarding this metal simply as a sort of malleable stone, he contented him self with obtaining pieces of suitable size from the ground or from natural blocks or veins, and hammer ing them into the desired shapes.
The Lake-Superior region furnished the Indians with most if not all the copper they used. The in teresting researches of Mr. Whittlesey ' and others have shown how extensive were those ancient mining
1 "Ancient Mining 00 the Sborcs of Lake Superior."--Smithsonian Contribu tions to Knowledge. April, 1863.
232
AI^TIQUITIKS OF THE SOTJTIIEKN INIHANS,
operations along the shores of that lake. The meth ods and. implements employed for detaching pieces of this metal from their natural masses, and hammering them into various shapes, have also been carefully noted and described. Messrs. Squier and Davis, 1 Pro fessor ^Wilson,2 and others, concur in the opinion that the copper used by the Indians in the preparation of these cold-wrought implements and ornaments was obtained chiefly, if not entirely, from the ancient mines of Lake Superior.
This copper axe from the shores of Lake Superior in the stone grave of an Indian in the beautiful valley of Nacoochee is surely an interesting proof of the com merce which existed among the tribes of North Amer ica. In this connection we would refer to another ornamental copper axe and to some copper rods or spindles found in an ancient grave in the Etowah Val ley. It will be remarked how closely this implement (Plate VI., Fig. 3) resembles the Oxaca axe figured by Du Paix. Like the thin axe from the stone grave in Nacoochee Valley it is of native copper, laminated in its structure, and "was hammered into its present-shape without the intervention of fire. This axe, also, is thin, and coiild not "well have been used for incisive purposes.
The design of the small copper rods, of "which Fig ures 4, 5, 6, and 7, Plate VI., are illustrations, it is dif ficult to conjecture. It has occurred to us that they may be the spindles alluded to by the historians of De Soto's expedition, with which, "when heated, the natives were wont to perforate pearls so that they could be strung and worn as beads.
1 " Ancient Monuments of the Mississippi Valley," p. 203. a " Prehistoric MaiV p. 174.
COPPER OP.NAMENTS.---- CASSIS FLAM3IEA.
233
Copper pendants for the ear have been found in several of the valleys of Upper Georgia. They were small, thin, pear-sliaped ornaments, perforated in the upper end to facilitate the suspension. The surfaces of these pendants -were frequently ornamented "with in cised lines, apparently traced with a flint flake.
Two fine specimens of the Oassis ftaimnea were taken from these graves--one of them from the central sarcophagus. They were nearly ten inches in length and about seven inches in diameter. From them, both the interior whorls and coluniellas had been removed, so that they answered the purpose of dririking-cups or receptacles of some sort. Dr. Troost saw in Tennes see one of these shells with an idol inside of it--an opening having been made for its reception. This im age was in a kneeling posture, "with its hands clasped in front. 1 Dr Drake found in some ancient tumuli near Cincinnati, large marine shells of the sort we are now considering. They had been cut longitudi nally so as to form very convenient drinking-cups.3 Professor Jones informs me that conchs of this de scription were seen by him not infrequently iu the stone graves of the Cumberland Valley. The pres ence of marine shells in graves of a similar character, in other localities, has been noted by more than one observer. These conchs -were brought from the South ern Atlantic coast, or from the shores of the Gulf of Mexico. Here we have another illustration of the commerce which must have obtained among the tribes
of this country. A soa/pstone ornament and several sliell pins were
obtained from the central grave. Of the shell pins or
1 Transactions American Ethnological Society, vol. i., p. 361. 1 Long's "Expedition to the Rocky Mountains," etc., vol. i., pp. 57, 58. Lon don, 1823.
234
ANTIQUITIES OF THE SOUTHERN INDIANS.
ornaments there are two varieties, both of which are correctly represented in the accompanying plate. The first (Fig. 8, Plate VI.) is like a large-headed wroughtnail. The other (Fig. 9, Plate VI.) is pointed at both ends. The surface of each is much decomposed, color, dead -white, and the feel, limy. The third relic (Fig. 10, Plate VI.) is made of soapstoue, and has received as careful a polish as that material will permit.
The precise uses to which these implements or ornaments may Lave been put can only be conjectured. That the form is not accidental is established by the coinciding shape of the soapstone specimen, and by the fact that several specimens of both varieties of the shell pins were found in the central grave. Similar rel ics have recently been taken from Brake-bill mound in Tennessee, and from a mound on the Chattahoochee River below Columbus. We will allude again to these articles in our chapter upon shell ornaments.
The soapstone ornament is a little less than two inches long. Its head is rather more than half an inch in diameter.
The shell pins with heads . are, on an average, about an inch and three-quarters in length. The mean diameter of their heads is three-quarters of an inch. They all terminate in a sharp point. Those without heads are fully two inches in length, swell ing in the centre to three-tenths of an inch in di ameter, and tapering to a point at either end. These shell ornaments were manufactured probably from the thicker portions and columns of sea-conchs or marine shells.
Cabeca de Vaca declares that the thicker portions of large marine shells, and of sea-conchs, were carried by the natives who occupied the coast-regions of the
1
ARTICLES FOUND IN STONE GKAV-ES,
235
Gulf of Mexico, into the interior, and were there ex changed for skins and other articles. In some such way, in all likelihood, was the material obtained from which these shell pins were fashioned.
The perforated stone 1 in its shape and size is not unlike the spindle-whorls found at Meilen and else where in Europe.1 Whether it -was indeed used for a similar purpose, or merely -worn as an ornament, or with -what specific intent fashioned, -we cannot state with, certainty. We incline to the belief that it was probably suspended as an. ornament.
An imperforate discoidal stone, a grooved axe, badly -worn, a beautifully-polished wedge-shaped axe or stone celt, a chisel of greenstone, a fragment of a soapstone pipe, and a large stone bead (Fig. 12, Plate VI.) complete the catalogue of relics taken from these graves.
In the vicinity -were ploughed tip Venetian beads, fifty-five in number, varying in shape and color, some of them being red, others blue, others white (of-which variety some have a blue -wreath, inlaid, encircling them), others green, "with crimson and yellow hori zontal stripes upon them, and others black. The ma terial of -which they are all made is either glass or por celain.
In the ancient town of Cutifachiijui, De Soto * found a dirk and beads which belonged to Euro peans "who, the Indians said, had many years before sailed into the port distant two clays' journey from this point.
Biedma * narrates that De Soto, while at this In-
1 Fig. 11, Plate VI. 2 See Relief's " Lake Dwellings of Switzerland," Plato III., Fig. 13. 3 " Relation of the Knight of 151 vas," p. 64 (Buckingham Smith's translation). 4 " Relation," etc., p. 240 (Buckingham Smith's translation).
236
ANTIQUITIES OF TITE SOUTIIERK INM>IANS.
dian village, caused a mosque to be opened in which "were interred the chief personages of the country. In it were found pearls, two "wooden axes of Castilian make, a rosary of jet beads, and some false pearls, such as were brought from Spain for the purpose of traffic with the natives. It was conjectured that these Euro pean articles had been obtained by the Indians from the followers of the Licentiate Ayllon.
Hosaries ' of glass beads, with crosses and hatchets of Vizcayan. make, were discovered in unrolling the bodies of some dead Indians entombed (if we correctly interpret the geography of the expedition) on the bank of the Savannah Hiver. This occurred in 1540, while the army of De Soto was enjoying the hospitali ties of the queen who ruled over the tribes of this region.
^Vhen speaking of the beads manufactured by the natives, the early historians enumerate such as were made of pearls, shells, sea-snails, stone, clay, and bone. From Indian mounds on the Georgia coast the writer has taken glass and porcelain beads, which proves that the custom of mound-building, or at least of interring the dead in mounds already constructed, existed at a period subsequent to the early intercourse between, the Indians and the Europeans. In an oval mound aboiit six miles, by water, above Lake Monroe, in Florida, Dr. Briuton saw numerous small blue and large "white glass beads, "which he regarded as inhumed at the time of the formation of the tumulus.2
"We know that the Spaniards brought quantities of European beads with them, with, which to conciliate the natives, and that the missionaries "who accompanied
1 Fontancda, p. 45 (Buckingham Smith's translation). s " Notes on the Floridian Peninsula," p. 170.
BEADS OF EUROPEAN MAHTJFACTTJKE.
237
their expeditions "were not "wanting in an abundant supply of rosaries. ISTothiiig -was more common, or, according to the report of the times, more conducive to the spread of Christianity in these benighted wilds, than the general dissemination of rosaries and wooden crosses. The acceptance at the hands of the priests, by the Indian, of such a gift, was too often chronicled as an instance of conversion.
Beads were also distributed by the early navigators at various points along the coast. The discoverers of the Mississippi dispersed them freely among the tribes then peopling the banks of the " Father of Waters," and -the wanderings of the pioneers of the west are still verified by the presence of these coarse orna ments. After having grossly violated the hospitali ties of the Queen of the Savannah, De Soto moved with his command along the line of the Savannah River to its head-waters. Thence turning to the south west, before reaching the confluence of the Etowah and the Oostenaula [Rivers, in his journeyings through Cherokee Georgia, it is probable that he passed either directly through or very near JVacoochee Valley. 1
Certain it is that, during the sixteenth century, ample opportunity would have been afforded to a prominent chief of this valley to have possessed him self of such beads as those "which "we are now ex amining.
No trace of iron, bronze, or steel, existed in these graves. The presence of the copper axe and stone im plements furnishes good ground for believing that their owner had enjoyed no opportunity for exchang-
238
ANTIQUITIES OF THE SOUTHERN" INDIANS.
ing his rude weapons and ornaments for the more ser viceable tools which, at an early period, were freely of fered by the colonists. This fact, and the total absence of the old Venetian beads 'found in the neighborhood, and undoubtedly once the property of the Indians, enable us, with considerable confidence, to assign to these graves an antiquity of not less than three hun dred and thirty years. Probably they are much older. This peculiar mode of sepulture, "we have already seen, was adopted by the Indians, in some instances, within historic times.
The existence of extensive trade relations among the aborigines is beautifully verified by the contents of these graves. Here, concentrated in the ownership of a single individual, we see a basket or mat made of a reed not native to the valley--stone implements, la boriously manufactured of materials brought from a distance--a cassis and shell ornaments from the Atlan tic Ocean, or the Gulf of Mexico, and a copper axe from the shores of Lake Superior.
Upon the rock-walls which fence in this beautiful valley, are engraven no memories of the tribes who in the olden time committed to the protection of these rud.e sepulchres the bodies of their dead.
-A. people without letters, they have passed away, leaving not even an inscription upon their tombs.
Nevertheless, from out the void of forgotten centu ries, from the womb of these nameless sarcophagi, come these implements and ornaments to tell us at least somewhat of the manners and customs of those who are gone--to remind us of the careful considera tion they bestowed upon the last resting-places of their departed--of their belief in a future state--of the prog ress they had made in the rudiments of art--of the
COMMERCE AMONG THE ABORIGINES.
239
position they occupied, in the scale of semi-civilization --and to assure us that among the red nomads of these primitive wilds the advantages of an interchange of values with distant nations were neither wholly un known nor entirely neglected.
Thus these relics become in very deed the
" Registers, the chronicles of the age They were made in, and speak the truth, of history Better than a hundred of your printed Communications." *
1 Shakerly's "Marmyon's Antiquary."
CHAPTEK XL
Arrow and Spear Heads.--Use of tlio Bow.--Skill in Archery.---MMaanmuiffaaetiure and General Distribution of Arrow and Spear Points.--Various Forms of- ti'n re Implements.--Stone Dagger.--Flint Sword.
OF all tlio various stone implements evidencing tlie handiwork of primitive man, by far the most numerous, and perhaps not tlie least interesting, are the arrow and spear heads. So general is the distribution of these in struments of war and of venery, not only throughout the length and breadth of vast continents, but also in the habitable islands of the ocean, it would really ap pear as though in every quarter of the globe, at some time or other, man existed in such a state of rude de velopment that his principal hope of food and defence resided in the constant use of these rough weapons. So closely do these implements resemble each other, both in material and form, of construction, whether found in Danish shell-mounds or British barrows, ex humed from the peat-bogs of Ireland, or wrested, from the diluvial matrix of France, brought to light from, out the darkness of long-forgotten caves, or fished up from the pile-dwellings of Hobenhauseri, gathered amid the forests of Africa, or upon the steppes of Asia, res cued from the debris of a ^New-Zealand encampment, or delivered from the womb of an American tumulus,
DISTBIBUTION OF' AKKOW AND SPEAB HEADS. 241
that even the practised eye is often at a loss to dis cover physical peculiarities which can sufficiently dis tinguish them the one from the other. Chronologically considered, the stone periods which they represent may Tbe separated by hundreds and perhaps thousands of years. ^Nearly synonymous as these relics appear, it Iby no means follows that they are necessarily synchro nous. The tendency of present investigations is to the conclusion that in America the stone age reached its richest and fullest development, and that the arrowmakers of the Western Hemisphere were surpassed by none in the selection of their materials and in the skill displayed in fashioning them into forms of use and symmetry.
AVithin the geographical limits of Greorgia spear and arrow points of unusual beauty and excellence are found in sepulchral tumuli, in shell-heaps, in. relic-beds, and in greater or fewer numbers upon the surface of the ground. New specimens are, each year, unearthed by the ploughshare and washed from their hidingplaces by the summer showers. Most frequently are they seen in the rich valleys, along the banks of rivers, and upon the islands and headlands of the coast. Their presence in quantities indicates the chosen seats of the aborigines. Even in pine-barren regions, where the soil is poor, vegetation thin, and streams and swamps are infrequent--localities at best but sparsely populated by the Indians--"we are assured of the fact that over these uninviting districts the natives wandered in pur suit of game, and here and there lost an arrow-point car ried away by a wounded animal or bird, broken from its shaft by contact with some, tree, or accidentally dropped from the hunter's quiver. Occasional specimens are turned up in ditching the rice-fields, showing that even
242
ABTIQUITIES OF TSK SOUTHERN INDIANS.
amid the deep, dank recesses of the primeval cypressforests the Indians pursued the deer, the bear, the wildturkey and other game. It is in the rich valleys of Middle and Upper Georgia, and along the coast, that these relics occur most frequently. Ready supply of oysters, clams, fresh-water mussels, fish, and game, mainly determined the natives in the selection of their permanent habitations. Add to this the presence of a good natural spring, and you will readily find abun dant proof of former occupancy by the red race. There can be no doubt but that the manufacture of these flint implements was carried on at various points oil the headlands and islands of the coast. In the refusepiles, numerous spear and arrow points may be gath ered, some completely and others only partially formed. Chips, flakes, and cores mark the spots where the primitive arrow-maker plied his trade, and this at a -considerable remove from localities whence the mate rials for the construction of these implements could have been obtained.
The writer has observed upon the margin of more than one swamp in Southern Georgia the clearly-de fined traces of open-air workshops for the manufacture of flint implements. Let one instance suffice: On Arcadia plantation, in Liberty County, near the edge of IVIidway swamp, is a little knoll whose top is littered with flakes, chips, and arrow and spear points in vari ous stages of completion. Some had evidently been discarded during the process of manufacture, upon the discovery of an unexpected defect in the material; while others, failing to yield the desired fracture, had been thrown aside as involving too great an expendi ture of labor. Fine nuclei of flint and quartz lay half embedded in the soil. These had been brought
MAIfUFACTURIEBS OF ABBOW AND SPF.AB HEADS. 243
from a distance. They could not have been obtained within a hundred miles of this locality, and, in all likelihood were procured at a remove much greater than that which "we have named. The spear and ar row heads of the coast are remarkable for the beauty of fche material from which they "were made, and for the skill displayed in their construction.
Particular attention was paid to the selection of the more attractive and bright-colored varieties of flint, jasper, and quartz. Many of these arrow and spear heads are beautiful, and may be justly regard ed as marvels of skill in flint-chipping. Some are serrated, and almost every known form finds here its type. "We have seen that at least some of these implements were here manufactured from nuclei, brought from a distance. It is probable, however, that most of them were obtained in a manufactured state from other localities. It is said that, among the In dians of Cherokee Georgia, in ancient times, were men who devoted their attention to the manufacture of spear and arrow heads, and other stone implements. As from time to time they accumulated a supply, they would leave their mountain-homes and visit the sea board and intermediate regions for the purpose of ex changing these implements for shells and various articles not readily obtainable in the localities "where they resided. These were usually old men, or persons who mingled not in the excitements of war and the chase. To them, "while engaged in these commercial pursuits, free passage was at all times granted. Their avocation was deemed honorable, and they themselves were welcomed wherever they appeared. If such was the case, we have here an interesting proof both of the trade relations existing among the aboriginal tribes
244
ANTIQUITIES OF THE SOUTHERN INDIANS.
and of the marked recognition, by an uncivilized race, of the claims of the manufacturer.' In the symmetry and beauty of these coast implements, "we perceive the good sense and precaution exhibited by the primitive merchantman in selecting such arrow and spear points as would most, surely commend themselves to the taste of his customers, and secure for him the most liberal returns of the articles highly esteemed among his own
people. In the relic-beds and shell-heaps along the banks
of fresh-water streams--marking the localities where the Indians resorted for fishing--many rudely formed slate and graywacke arrow and spear points are found. There, the material being abundant and easily "worked, no special care seems to have been bestowed upon the manufacture of these implements. Jasper and milky quartz were also freely used. From some of these relic-beds thousands of these arrow and spear heads have been obtained, and in every stage of develop ment, from the rough stone just beginning to be worked, to the finished implement. In the valleys of Tipper Georgia our search has been rewarded by ex quisite specimens made of pellncid crystals, violet and smoky quartz, chalcedony, jasper, and flint. Party-col ored materials were evidently held in special esteem, and so neatly are many of them chipped that the skilled lapidary of the present day might find his powers taxed to rival the workmanship here dis
played. Absolutely dependent as the Indians were upon
the use of the bow and arrow and the spear for sub sistence and protection, nomadic in their habits, and
'"Indian Kemains in Southern Georgia," by Charles 0. Jonea, Jr., p. 19. Savannah, 1859.
SKILL 1ST AEOHEET.
245
constantly engaged ill hunting: or intertribal conten tions,' we are not surprised at tlie general distribution, all over the. face of the country, of these their simplest and most common implements. Their bows--long since unstrung--have crumbled into dust. .Arrowshaft and spear-handle are seen 110 more, but these nu clei, flint-chips, and skilfully-formed stone points-- indestructible by time--fortunately still remain for our study and information.
According to the relation of Cabeca de Vaca, the Indians were all archers, being admirable in their pro portions, spare, and of great activity and strength. Their bows were as thick as a man's arm, eleven or twelve palms in length, and capable of projecting arrows for a distance of two hundred paces, with such precision, as to miss nothing. Detailing a skirmish which occurred between the Spaniards and the natives, he says the latter fought from behind, trees, covering themselves so that the Christians could not get sight of them. He affirms that they " drove their arrows with such effect that they wrotmded many men and horses." Even the " good armor " of the Spaniards did not avail for their protection against these missiles. Some of the soldiers swore " that they had seen two redoaks, each the thickness of the lower part of the leg, pierced through from side to side by arrows. . . . This," continues the historian, " is not so much to be won dered at, considering the power and skill with which the Indians are able to project them. I myself saw an
1 Wilson, writing in 1682, asserts that the Carolina tribes wore so constantly engaged in wars " one town or village against another," that the Indian popula tion suffered " no increase of People "--several nations having been " in a manner quite extirpated by Wars amongst themselves since the English settled at Asftty River." " Account of the Province of Carolina," p. 15. London, 1682.
246
ANTIQUITIES OP THE SOUTHEKjM INDtAlfS.
arrow that had entered the Ttmtt of an elm to the depth of a span."
Oil another occasion the same author states that upon an examination of the corpses of the Spaniards "who had fallen in battle, their foodies were found to have been traversed from side to side by arrows; and although some of the dead soldiers were clad in good armor, it did not afford adequate protection or security against the nice and powerful archery of the Indians. An instance is given where an arrow, shot by an Indian, pierced through the saddle and housings and penetrated one-third of its length into the body of of a Spaniard's horse. The bows were made of wood and remained unbent until needed for battle or the chase. The strings were formed of deer-sinews, strips of deer-skins, or the twisted gut of animals. Hard canes and "wood were used for arrows. 1
The Fidalgo of Elvas describes the Indians as be ing exceedingly ready with their weapons, and " so warlike and nimble that they have no fear of footsoldiers ; for if these charge them they flee, and when they turn their backs they are presently upon them. They avoid nothing more easily than the flight of an arrow. They never remain quiet, but are continually running, traversing from place to place, so that neither cross-bow nor arquebuse can be aimed at them. Be fore a Christian can make a single shot with either, an Indian will discharge three or four arrows : and he sel dom misses of his object. When the arrow meets with no armor, it pierces as deeply as the shaft from a cross-bow. Their bows are very perfect; the arrows are made of certain canes, like reeds, very heavy and
1 See "Relation Qf AWar Nuney, Caboea <3e VacEi," translated from the Span ish by Buckingham Smith, pp. 39, 40, 48, et seq. New York, 1871.
TISE OF ABEOWS.
.
247
so stiff that one of them, when sharpened, will pass through a target. Some are pointed with the bone of a fish, sharp, and like a chisel, others with some stone like a point of diamond; of such the greater number, when they strike upon armor, break at the place the parts are put together; those of cane split and will enter a shirt of mail, doing more injury than when armed (i. e., .with chips of hard stone). 1
" The Indians," continues the same narrator, " never lack meat. With arrows they get abundance of deer, turkeys, conies, and other -wild animals, being very skillful in killing game."
While the army of De Soto was quartered in the province of Cofachiqui, A.fiasco was dispatched by the governor to secure the attendance of the mother of the princess of this country, who was represented to be in the possession of a large quantity of most valuable pearls. As a guide he took with him a youthful war rior, brave and handsome, and a near relative of the princess. Having proceeded nearly thres leagues, Anasco and his comrades halted for their mid-day meal. While they were reposing beneath the shade of some wide-spreading trees, the Indian guide, who had become very moody and thoughtful, quietly took off his quiver, and, placing it before him, drew out the arrows slowly, one by one. They "were admirable for the skill and elegance with which they were formed. Their shafts "were reeds. Some -were tipped -with bucks'-horns, -wrought with four corners, like a dia mond ; some were pointed "with the bones of fishes, curiously fashioned.; others -with barbs of the palm
248 AjrriQuntES OF THE SOUTHERN ISDIAKS.
and other hard woods; and. some "were three-pronged. They were feathered in a triangular manner to render their flight of greater accuracy. The Spaniards could not sufficiently admire their "beauty; they took them up, and passed them from hand to hand, examining and praising their workmanship and extolling the skill of their owner. The youthful Indian continued thoughtfully emptying his quiver until, almost, at the last, he drew forth an arrow with a point of flint, long and sharp and shaped like a dagger; then, casting around a glance, and seeing the Spaniards engaged in admiring his darts, he suddenly plunged the weapon in Ms throat and fell dead on the spot.
TJnwilling to Betray the place of concealment of his mother, and fearing to incur the displeasure of Ms queen by disobedience of her mandate, he gave him self willingly to death. 1
The Chevalier Tonti,' alluding to the force with which their arrows were projected by the natives, says: " That which is wonderful in this, is the havock which the Shot sent by the Savages makes; for, besides the e'xactness and swiftness of the Stroke, the force of it is- very surprizing, and so much the rather, because it is nothing else but a Stone, or a Bone, or sometimes a piece of very hard Wood pointed and fastned to the end of an Arrow with some Fishes-glue, that causes this terrible effect. AVlien the Savages go to "War, they poison the Point or extremity of their Dart so that if that remains in the Body Death follows of necessity; the only Remedy in this case is to draw out the Arrow through the other side of the "Wound, if it goes quite
1 Irving's " Conquest of Florida," chapter xlvii., pp. 225, 226. New York, 1851.
3 " An Account of Monsieur do La Sallc's Last Expedition and Discoveries in North America," p. 71. London, 1698.
ABROWS AND AKBOW--POINTS.
249
through; or, if not, to make an aperture on the other side,. and so to draw it through; after which they know by instinct certain Herbs the application of which both draws out the Venom and Cures them."
In Laudonniere's introduction to his history of Jean Ribault's .first voyage to Florida,' we are told that the Indians had no weapons other than bows and arrows. The bow-string was made " of the gut of the stag, or of a stag's skin which they know how to dress as -well as any man in France, and with as different sorts of colors. They head their arrows with the teeth of fishes, and stone -which they work very finely and handsomely."
The following interesting account is taken from the " History of the Bucaniers of America:" ' " On the ninth day after our arrival, our women slaves being busied in ordinary employments of washing of dishes, sewing, drawing water out of -wells which we had made oil the shore, and the like, one of them -who had seen a troop of Indians towards the woods, cried out Indians, Indians I "We ran presently to our arms and their re lief, but coming to the wood we found no person there, but two of our "women slaves killed, upon the place with arrows: in their bodies we saw so many arrows stick ing, as if they had been fixed there with particular care, for otherwise we know that one of them was sufficient to kill any man. These arrows -were all of a rare shape, being eight feet long, and as thick as a man's thumb; at one end was a hook of wood, tied to the body of the arrow with a string, at the other end was a case or box like the case of a pair of tweezers, in
1 " Historical Collections of Louisiana and Florida," by B. F. French. New Series, pp. 170, 171. New York, 1869.
3 Vol. i., pp. 212, 218, fifth edition. London, 1771.
250
ANTIQUITIES OF THE SOUTHERN" INDIANS.
which we found little pebbles or stones ; the color was very red, very shining, as if they had been locked up, all which we believed were the arms of their leaders. These arrows were all made without instruments of iron ; for whatever the Indians make, they harden first artificially with fire, and then polish them with flints." To Thomas Ash we are indebted for the following de scription of the Carolina Indians, as he saw them in 1682 : 1 "The natives of the country are from time im memorial ob Origine Indians, of a deep Chestnut colour, their Hair black and streight, tied various ways, some times oyl'd and painted, stuck through with Feathers for Ornament or Gallantry ; their Eyes black and spar kling, little or no Hair on their Chins, -well limb'd and featured, painting their Faces with different Figures of a red or sanguine Colour, whether for Beauty or to ren der themselves formidable to their Enemies, I could not learn. They are excellent Hunters ; their Weapons, the Bow and Arrow, made of a Kead, pointed with sharp Stones or Fish-Bones; their Cloathing, Skins of the Bear and Deer, the Skin drest after their Country Fashion."
Discussing the " handicrafts" of the Virginia In dians, Beverly writes : " Before I finish my account of the Indians it will not be amiss to inform you that when the English went first among them, they had no sort of Iron, or Steel Instruments; but their Knives were either Sharpen'cl Keeds or Shells, and their Axes sharp Stones bound to the end of a Stick and glued in with Turpentine. By the help of these they made their Bows of the Locust Tree, an excessive hard Wood when it is dry, bnt much more easily cut when it is green, of which they always took the advantage. They
1 " Carolina ; or a Description of the Present State of that Country," etc., pub lished by T. A., Gent., pp. 34, 35. London, 1682.
AKROWS, HOW MADE AXD AHMEL).
2ol
made their Arrows of Reeds or small "Wands, which needed no other cutting but in the length, being other wise ready for Notching, Feathering, and Heading. They fledged their Arrows with Turkey Feathers, which they fastiied ~\vith Glue made of the Velvet Horns of a Deer, but it has not that quality it's said to have, of holding against all Weathers; they arm'd the Heads with a white transparent Stone, like that of Mexico mention'd by Peter Martyr, of which they have many [Rocks ; they also headed them with the Spurs of the Wild Turkey Cock." '
Adair testifies to the accuracy with which the Cherokee Indians, in his clay, used their bows and ar rows arid threw their feathered darts. Speaking of these peoples, he declares: " They make perhaps the finest bows and the smoothest barbed arrows of all mankind. On the point of them is fixed either a scooped point of buck-horn, or turkey-cock spurs, pieces of brass, or flint stone. The latter sort our forefathers used, which our witty grandmothers call elf-stones, and now rub the cows with, that are so unlucky as to be shot by night fairies. One of those flint arrow-points is reckoned a very extraordinary blessing in a whole neighborhood of old. women, both for the former cure, as well as a preservative against every kind of bewitch ing charm." "
As early as 1761 the Cherokees seem to have aban doned the use of stone arrow-heads, and in their stead to have substituted points of metal. Lieutenant Henry Timberlake, from his personal observations, furnishes us with their method, then in vogue, of pointing arrows : " Cutting a bit of thin brass, copper, bone, or scales of
1 " History and Pi-esent State of Virginia," boob iii., p. 60. London, 17O5. 2 " History of the American Indians," p. 425. London, 1775.
I
252
ANTIQUITIES OF THE SOUTHERN INDIAN'S.
a particular fist, into a point with two beards, or some into an acute triangle, they split a little of their arrow, whieli is generally of reeds; into this they put the point, winding some deer's sinew round the arrow, and through, a little, liole they mate in the head ; then they moisten the sinew with their spittle, which, when dry, remains fast glewed, nor ever untwists. Their bows are of several sorts of wood, dipped in bear's oil, and seasoned before the fire, and a twisted bear's gut for the string." *
Spear-heads were fastened to wooden handles, eight or ten feet in length, and were hurled as javelins. At close quarters they "were employed to ward off blows, and to d.eliver thrusts, -without quitting the hand. A.mong the Carolina Indians Lawson observed in 1V01 long arrows headed with pieces of glass, which they had broken from bottles. " They had shaped them neatly, like the head of a dart, but which way they did it I cant tell." * It is not improbable that this historian mis"took the material of which these ' arrow-points were made. The resemblance between some varieties of quartz, or obsidian and glass, is so close, than an error may thus have occurred in the ob servation. We know that the Indians of California sometimes make arrow-heads from old glass bottles, and Captain Cook states that the New-Zealanders found means to drill a hole, -with jasper, through a piece of glass which he had given them, so that it might be suspended as ail ornament from the neck. It may be, therefore, that the remark of Mr. Lawson is entirely 'correct.
"Without multiplying these historical references, it
> " Memoirs of Lieutenant Henry Timberlake," pp. 61, 62. London, 1Y65. 3 Lawson's "History of Carolina." Reprint, p. 99. Raleigh, 1860.
AM. fWOTO-UrHOGRAfVK CO-N.f.fOSBOfiNfS P/
LABGE SPEAK--HEAD.
253
is evident that, at the period of our earliest acquaint ance with the Southern Indians, tie bow and arrow were in general use, constituting, in the hands of the natives, an indispensable, effective, and deadly -weapon. Appreciating the vast numbers of arrow and spear heads which, during the lapse of many centuries, must nave been manufactured and expended by these peo ples in hunting, fishing, in their games and in frequent wars, ~we are prepared fully to understand why these flint implements are found in such quantities, and why they should form the most common proof of the former occupancy of the soil by the red-men.
The largest spear or lance head we have seen within the geographical limits of Georgia, was obtained in a grave-mound -which stood upon the point of land formed by the confluence of the Etowah. and Oostenaula Rivers. It is nearly fourteen inches in length, and three inches and a quarter in -width--weighing two pounds and two ounces, avoirdupois. (^Sfe^Fig. l,jPlate VII.) It is perfect, -with the exception of the point which, "was broken off at the time this implement was taken from the mound. No spear-head of such magni tude, so far as my knowledge extends, has been found within the limits of the Southern States. It is made of flint, and the conchoidal fractures caused in removing the flakes are clearly denned. The tumulus from which, this spear-head was obtained -was circular in shape, about twelve feet high, and -with a base-diameter of fifty feet. It contained numerous skeletons, and afforded a rich yield of various and interesting relics. Sharing the fate which has overtaken so many of these aboriginal monuments, but little now remains to mark the spot once rendered so attractive by the presence of this beautiful tumulus. The grand forest-trees which
254
ANTIQUITIES OF THE SOUTHERN INDIANS.
formerly grew upon and threw their protecting shad ows about it--trees which, in all likelihood, sheltered De Soto and his companions, as resting upon the ver dant banks of the Etowah they were hospitably enter tained by the Cacique of Ichiaha--have all been cnt down, and the earth and clay composing the mound carted away to assist in levelling the streets of the city of Rome, and aid in the construction of a landingplace for a ferry-boat.
The second spear-head (Fig. 2, Plate VII.), pointed at both ends, and regularly chipped, was found in the valley of the Chattahoochee, a few miles below the city of Columbus. It is twelve inches and a half in length, two inches and four-tenths wide, and weighs ten and three-quarter ounces. It was probably hafted in a bone, horn, or wooden socket at the end of the shaft. The similarity bet-ween this implement and that figured by Messrs. Squier .and Davis, on page 211, of the " A-ncieiit Monuments of the Mississippi Val ley," is remarkable. I refer to No. 3 in ITig. 99. It may be that this formidable implement was used as a dagger.
The "beautiful spear-heads represented by ISTos. 1 arid 2, Plate VIII., were taken from a chieftain mound near Darien, in Macintosh County. The remaining figures in this plate illustrate the prevailing types of these implements, as they are to this day found in tumuli, ploughed up in the fields, gathered from relic-beds, or picked up on the sites of ancient villages and open-air workshops.
It will be perceived that among the arrow-heads figured in the accompanying plate, 1 are all the varieties enumerated by Sir W. E. Wilde, viz., the triangular,
1 Figs'. 1 to 41 inclusive, Plate IX.
TYPICAL FORMS OJ? AKROW--POINTS.
255
the indented, the stemmed, -with a tang or projection, for insertion into the shaft--the barbed and the leafshaped. Nor does this catalogue embrace them all. The modifications of the one idea of arming the point of the reed or wooden arrow with a piece of chipped flint are numerous. From a collection of more than two thousand, we have selected these as presenting those forms in general use. The shark's tooth may have suggested the shape of the indented arrow-point, and the serration of the edges. Fossil-shark's teeth are found in various parts of Georgia and Carolina. Their existence was known to the aborigines, who some times perforated and wore them as ornaments about their necks. The writer lias taken them from the earth-mounds on the coast.
Some of these arrow-points (Fig. 40, Plate IX.) are flat, -with their edges bevelled in opposite direc tions. The object of this arrangement was to cause the arrow, in its flight, to take a rotary motion, thereby increasing the violence of the wound when the barb had entered the flesh. The same effect was accom plished by using the half twist in feathering the shaft. By means of such mountings the flight of the arrow Avas rendered more steady.
The use, by the Indians, of the hard canes, so com mon in the Southern swamps, as arrow-stems, seems to bo substantiated by the spike-shaped flint tips of which ^To. 32, Plate IX., may be regarded as typical. These could readily have been inserted in the hollow of the reed, cut for that purpose at such a remove from the joint nearest the butt of .the arrow, that the inserted end of the spike, closely fitting, would rest against and be held in position by it. Often has the writer adopted this method of spiking his arrows, during his
256
ANTIQUITIES OP THE SOUTHERN INDIANS.
youthful days, and with flint tips precisely like .those under consideration, gathered from the shell-mounds and picked up in the old Indian fields on the Colonel's Island.
The taste displayed in the selection of choice and "beautiful material, and the skill exhibited in the man ufacture of these spear and arrow heads are often sur prising and excite our admiration. Pellucid crystals, smoky quartz, chalcedony, carnelian, jasper of varie gated coloring, flint, and other hard stones, were chipped with a regularity and delicacy truly astonish ing. Especially does our "wonder grow when we re member that these results were accomplished -without the aid of metallic tools. Perfectly-formed arrowpoints exist in the writer's cabinet, less than half an inch in length. 1
Bossu says that, among the Chactaws, the children exercised themselves in shooting -with a bow and arrow for prizes. " He that shoots best gets the prize of praise from an old man, who calls him an apprentice warrior; thus they are formed by emulation, -without corporal punishment; they are very expert in shooting with an instrument made of reeds, about seven feet long, into -which they put a little arrow, feathered with the wool of a thistle, and, in aiming at an. object, they blow into the tube, and often hit the aim, and fre quently kill little birds -with it.""
Captain Homans, in speaking of the same Indians, writes: " The young savages also use a very strait cane,
1 Lieutenant Timberlake, in his memoirs (p. 45), mentions the circumstance that the Cherokee children at eight and ten years of age were very expert at killing birds and small game,-with a sarbacan or hollow cane through which they blew a small dart, and with such precision that they rarely missed of striking "the larger sort of prey " in the eye. These small arrow-points, to which allusion has been made, were probably chipped for children's arrows.
a " Travels through Louisiana," vol. i., p. 306. London, 1771.
BLOW--GUNS.----BOWS AMD ABROWS.
2l
eight or nine feet long, cleared of its inward divisiotis of the joints; in this tliey put a small arrow, whose one end is covered one-third of the whole length with cotton, or something similar to it; this they hold near est their mouth, and blow it so expertly as seldom to miss a mark fifteen or twenty yards off, and that so violently as to kill squirrels and birds therewith." *
Between this miniature arrow-tip and the large flint spear-head, measuring nearly fourteen inches, these implements are seen of every intermediate length, and of various colors--red, bine, yellow, white, black and brown predominating. The arrow and spear points of the Southern Indians, as a gen eral rule, are more beautiful than those manufac tured by tribes who inhabited northern latitudes. The abundance of birds and small game in the swamps and deep forests of this semi-tropical region invited the use of flint implements of a delicate character.
If not inserted in the end of a hard cane, the arrowpoint was attached to a reed or -wooden shaft (a slit or notch having been made for its reception) by means of moistened threads of deer-sinews, glue, or small strips of buckskin. The moistened fibres of deersinews were generally used. These filled all inequalities, both in the stone tip and in the butt of the ar row, were very tenacious, and when dry compassed the juncture auite securely. Hickory, locust, white-oak, ash and red-cedar are said to have been the favorite woods employed by the Indians in the manufacture of their bows. These they seasoned thoroughly by artificial heat, and frequently anointed with bear's grease to render them flexible and keep them from
1 " Concise Natural Historv of East and West Florida," etc. New York. I'JYS,
'258
ANTIQUITIES OF THE SOUTHERN INDIAN'S.
cracking or breaking. The customary shape of the bow was that of a single curve, strengthened, in the middle. Usually plain, these bows were sometimes ornamented, their ends terminating in tips of bone or stag's horn.
De Bry * has favored us with several representa tions of the bows, arrows, and quivers of the Florida Indians.
His general description is conveyed in. the follow ing words: . . . " strenui tameu & pugnaces, nee alia habeiit, prseter areum & sagittas, arma. Arcus nervuiii ex cervino intestine aut corio adeo concinne pararc norunt, ut GalH melius jion possint, cfc variis coloribus inficiunt; pro mncroiie sagittarum sunt pisciam dentes <fo lapides affabre adaptati. A.dolesceiites cursu, sagittarum missioiie & pilse ludo exerceutur. . . . Venatione & piscatione magnopere delectantur." '
Of the "weapons of the Virginia Indians he says : " Eorum arma ad nocendum sniit dtintaxafc arctis ex corylo, & sagittao ex arundine, cleinde stipites lignei plani, duorum cubitoruni longitrtdiiie ; " and furnishes us in plate iii., as well as elsewhere, with the form of the bow, arrow, and quiver. 3
Reserve arrows were carried in a rude quiver, made of deer, fawn, or cougar skin, suspended from the left shoulder, and hanging just behind the right hip where most convenient access could be had. Often these arrows were not feathered, tlie weight of the stone tip being of itself sufficient to preserve regularity
" Admirauda Xorn D 1600.
FOBCE WITH WHICH AKKOWS "WEED PROJECTED. 259
of flight. Tile rapidity and pi'ecision with which tlio Indians discharged tlieir arrows, when occasion re quired, are emphatically testified to by the Spaniards, who in those early days, because of their cruelties, incurred the enmity of the natives.
In the battle of Manilla there fell, of the armor-clad. Christians, two hundred. Of the living, one hundred and fifty received seven hundred wounds from the Indian arrows. Here is proof most emphatic of liow valiantly and. successfully the red-men, could handle their inide weapons, in. the face of mailed warriors, in. defence of home and country. Testimony is not want ing substantiating the efficiency and force with, which the arrow is projected by modern Indians. The history of the early conflicts of the colonists is filled with ex amples of the deadly effects of such ancient artillery, and the Dakota chief Wah-na-tah is said, on one occasion, to have discharged his arrow with so much vigor that it entirely traversed the body of a female buffalo and killed her calf on the other side.1
It only remains for us to consider the method adopted by the natives in the manufacture of these flint implements. So far as my knowledge extends, the use of iron was entirely unknown among the primitive peoples of this region. Copper implements there were, of limited variety, but these occur very rarely, and aa'e too soft for contact with stone. ^Ve are consequently compelled to the belief that the Indians fashioned these spear and arrow heads by chipping them with implements of stone. It may be that the serrated edges, and perhaps some of the more delicate arrow-heads, were formed with the aid of instruments
1 Schoolcraft's " Archives of Aboriginal Knowledge," vol. iv., pp. 95, 9G. Philadelphia, 1860.
260
ANTIQUITIES OF THE SOUTHERN INDIANS.
of bone, ivory, or horn, as in the case of those manu factured by the western Esquimaux tribes, as described by Admiral SirE. Belcher. 1 This, however, is scarcely probable, although Captain John Smith, in his sixth voyage, speaking of the Virginia Indians, says : " His arrow-head lie quickly maketh with a little bone which he ever weareth at his bracept, of a splint of a stone or glasse, ill the form of a heart, and these they glew to the end of their arrowes. "With the sinews of deer and the tops of deer's horns boiled to a jelly they make a glue which will not dissolve in cold water."
"We are of opinion that the Southern Indians flaked their flint implements by percussion and not by press ure. The latter method might answer with obsidian, but it would prove an endless and futile process if quartz, chert, jasper, and flint are the materials used in the manufacture.
Schoolcraft thus describes the mode observed by !North American Ind.ia.ns in the preparation of flint arrow and spear heads : " The skill displayed in this art, as it is exhibited by the tribes of the entire continent, has excited admiration. The material em ployed is generally some form of horn-stone, sometimes passing into flint. This mineral is often called chert by the English mineralogists. No specimens have, however, been observed where the substance is gunflint. The horn-stone is less hard than common quartz, and can readily be broken by contact with the latter. Experience has taught the Indian that some varieties of horn-stone are less easily and regularly fractured than others, and that the tendency to a conchoidal fVacture is to be relied on in the softer varieties. It
MASTZFACTITBE OF AEBOW-POISTTS.
261
has also shown him that the weathered or surface fragments are harder and less'manageable than those quarried from the rooks or mountains. To break them he seats himself on the ground, and holds the lump on one of his thighs, interposing some hard sub stance below it. "When the blow is given, there is a sufficient yielding in the piece to be fractured, not to endanger its being shivered into fragments. Many are, however, lost. After the lump has been broken transversely, it requires great skill and patience to chip the edges. Such is the art required in this busi ness, both in selecting and fracturing the stones, that it is found to be the employment of particular men, generally old men who are laid aside from hunting, to make arrow and spear heads." *
Catlin, in his " Last gambles amongst the Indians," speaking of arrow-making among the A.paches, says: " Every tribe has its factory in which these arrow heads are made, and in those only certain adepts are able or allowed to make them for the use of the tribe. Erratic bowlders of flint are collected (and sometimes brought an immense distance) and broken with a sort of sledge-hammer, made of a rounded pebble of hornstone set in a twisted withe holding the stone and forming a handle. The flint, at the indiscriminate blows of the sledge, is broken into a hundred pieces, and such flakes selected as from the angles of their fracture and thickness will answer as the basis of an arrow-head.
" The master-workman, seated on the ground, lays one of these flakes on the palm of his left hand, hold ing it firmly down with two or more fingers of the same hand, and with his right hand, between the
1 " Archives of Aboriginal Knowledge,1 ' vol. iii., p. 467. rtiiladolphia, 1860.
262
AKTIQTJITIKS OF THE SOUTHERN" INDIANS.
thumb and two forefingers, places Ms chisel (or punch) 011 the point that is to1 be broken off; and a cooperator (a striker) sitting in front of him, with a mallet of very hard wood, strikes the chisel (or punch) on the upper end, flaking- the flint off on the under side, be low each projecting point that is struck. The flint is' then turned and chipped in the same manner from the opposite side, and so turned and chipped until the re quired shape and dimensions are obtained--all the fractures being made on the palm of the hand.
"In selecting a flake for the arrow-head, a nice judg ment must be used, or the attempt will fail; a flake with two opposite parallel, or nearly parallel planes, is found, and of the thickness required for the centre of the arrow-point. The first chipping reaches near to the centre of these planes, but without quite breaking it away, and each chipping- is shorter and shorter until the shape and the edge of the arrow-head are formed.
" The yielding elasticity of the palm of the hand enables the chip to come off "without breaking the body of the flint, which would, be the case if they were broken on a hard substance. These people haw no metallic instruments to work -with, and the instrument (punch) which they use, I was told, was a piece of bone ; b>ut on examining it, I found it to be a substance much harder, made of the tooth (incisor) of the spermwhale, which cetaceans are often stranded on the coast of the Pacific. This punch- is about six or seven inches in length, arid one inch in diameter, with one rounded side, and two plane sides, therefore presenting one acute and two obtuse angles to suit the points to be broken. This operation is very curious, both the holder and the striker singing, and the strokes of the mallet given exactly in time with the music, and with
MANITFACTUKE OF ABBOW--POINTS.
263
a sharp and rebounding blow, in which, the Indians tell us, is the great 'medicine (01* mystery) of the operation." 1
Commenting upon this description, Mr. Steveiis observes: " "What Catlin has said with reference to a rebounding blow, is perfectly true ; it is impossible to flake flint with a dull, heavy, smashing blow; it is the measured and rebounding blow--a shock rather than a blow--which, given with judgment, enables the mate rial to take its own line of cleavage, and produces what is so well known as the conchoidal fracture. It is the presence of this conchoidal fracture, resulting from hurnaii skill, that distinguishes the mere splinter of flint from the flint-flake." '
Ijietitenant Beckwith, in 1854, saw a Pah-Utah In dian, seated on the ground, make from a fragment of quartz, with a piece of round bone, one end of which was semispherical with a small crease in it (as if worn by a thread) the sixteenth of an inch deep, an arrow head which was very sharp and piercing, and in all respects similar to those in general use among the In dians of that region. He says : " The skill and rapidity with which it was made, without a blow, but by sim ply breaking the sharp edges with the creased bone, by the strength of his hands--for the crease merely served to prevent the instrument from slipping, afford ing no leverage--were remarkable." "
In 1860, Hon. Caleb Lyon communicated to the American Ethnological Society an account of the manufacture of arrow-heads of flint, glass, obsidian, and other materials, by the Shasta Indians of Califor nia : " The Shasta Indian seated himself on the floor,
264
ANTIQUITIES OF THE SOUTHERN" INDIANS.
and, placing the stoiio anvil upon Ms knee, winch was of compact talcose slate, with one blow of tiis agate chisel lie separated the obsidian pebble into two parts, then giving another blow to the fractured side, he split off a slab a fourth of an inch in thickness. Holding the piece against the anvil with the thumb and finger of his left hand, he commenced a series of continuous blows, every one of which chipped off fragments of the brittle substance. It gradually assumed the required shape. After finishing the base of the arrow-head (the whole being only little over an inch ill length), he began striking gentler blows, every one of which I ex pected would break it into pieces. Yet, such was their adroit application, his skill and dexterity, that in little over an hour he produced a perfect obsidian arrow head. I then requested him to carve me one from the remains of a broken porter-bottle, which (after two failures), he succeeded in doing. He gave as a reason for his ill-success, he did not understand the grain of the glass. No sculptor ever handled a chisel with greater precision, or more carefully measured the weight and effect of every blow, than this ingenious Indian; for, even among them, arrow-making is a distinct trade or profession, which many attempt, but in which few attain excellence. He understood the capacity of the material he wrought, and, before striking the first blow, by surveying the pebble, he could judge of its availability as well as tho sculptor judges of the perfec tion of a block of Parian. In a moment, all that I had read upon this subject, \vritten by learned and specula tive anticpuarians, of the hardening of copper for the working of flint axes, spears, chisels, and arrow-heads, vanished before the simplest mechanical process. I felt
DIFFERENT FORMS OF ARROW--POINTS.
265
that the world had been better served had they driven the pen less and the plough more." *
In view of these positive observations, it is fair to presume that the method adopted by the modern Indians in the manufacture of their common flint arrow and spear heads was but the perpetuation of a mode which existed among the red-men prior to historic times. It is the writer's impression that the flint implements found in Georgia and the Southern States were made by percussion--hammers of wood and stone, and stone chisels being used in removing the flakes.
In conclusion, it may not be uninteresting to analyze for a moment the prevailing types of these arrow and spear points. The primary, rudimentary, or simplest shape is that of either an isosceles or equilateral triangle (Figs. 1 and 2, Plate IX.).
How various soever the forms may be, upon exam ination they will be found to be modifications of this idea. Thus, if the lower corners of the triangular arrow-point are rounded, wo have the leaf-shaped implement (Fig. 3, Plate IX.).
Still preserving the triangular form, and merely cutting a notch on each side to facilitate its attachment to the shaft, we obtain the very common variety indi cated in Figs. 9, 15, and 21, Plate IX.
Hollowing out the base of the triangle gives us the indented or shark's-tooth form (Figs. 10, 26, 27, and 28, Plate IX.).
Add the notch on each side, and we see the beauti ful implement to this clay manufactured by the Cali fornia Indians, so skilfully, out of obsidian (Fig. 41, Plate IX.).
1 " Bulletin of the American Ethnological Society," vol. i., p. 39. New York, 1860-'63.
266
ANTIQUITIES OF THE SOUTHERN INDIANS.
The spike-shaped arrow-point (Figs. 31, 32, Plate IX.), found among the refuse-piles marking the spots where the Southern Indiana congregated for the pur poses of hunting and fishing, is but another modifica tion of the same triangular idea, to facilitate its inser tion in the hollow of the hard cane.
Various are the modifications of the base of the triangle, designed either to form a tang or projection for sinking into the shaft, or to facilitate its attachment to the arrow-stem.
Another form not infrequent in Georgia, and quite common in Tennessee, in which the apex of the triangle and the base have "both been rounded, is seen in Fig. 20, Plate IX.
To these may be added the arrow-point with a bifurcated tang (Fig. 36, Plate IX.).
"We ought to mention also the chisel-ended, the onebarbed or single-winged, and the repainted arrow-heads. M^e are inclined, however, to regard the numerous speci mens of these sorts which have passed under our obser vation (and of which "we have many in our collection) as examples rather of misfortune than of original design. They may Tbe rated as abnormal types, and reckoned as unwilling deviations, on the part of the manufac turer, from the symmetrical forms he desired to attain. Accident in manufacture, and the effort to remodel the implement after it had been broken, gave rise to most of these unusual varieties. They show how carefully these primitive peoples economized their stone weap ons, reforming them after they had been, seriously im paired, and using them even when they scarcely answered the accurate purposes for which they were designed. Under the general term wasters, we might enumerate many partially-formed, defective, and mis-
CLASSIFICATION OF ABROW--POINTS, ETC.
267
shapen arrow and. spear points with which the relicbeds and open-air workshops, located, upon the banks of many Southern streams, abound.
It is hardly proper, however, to pursue this attempt at classification any further. Were we to note all the varieties which suggest themselves, we would be led into a multiplicity of illustrations which would, do little more than represent the individual skill and fancies of the respective workmen, the various casual ties to -which these implements had been subjected dur ing the process of manufacture and subsequent use, and the modifications of form consequent thereupon.
Fashioned all after the same general idea, there is, nevertheless, in the many beautiful varieties which we encounter, in the delicacy and regularity with "which these flint implements have been chipped into forms of ornament and use, much to engage our attention and elevate our conception of the skill of these primi tive arrow-makers.
Before dismissing the further consideration of these implements of war, venery, and piscary, we would, refer to two unusual relics, one found in a grave-mound near the Warrior River, in Alabama, and the other taken by Professor Joseph Jones from a stone grave in Tennessee.
The former (Fig. 3, Plate VII.) is a flint dagger, well chipped, and seven inches and a half in length. If our information be correct, this is the first relic of this description which has been brought to light within, the territorial limits once occupied by the Southern tribes. In regularity of outline, and ex cellence of manufacture, it is not inferior to the Danish daggers--the handle being more completely formed than that of any of the three figured by Sir
268
ANTIQUITIES OF THE SOUTHERN INDIANS.
John Lubtooek on page 97 of his "Pre-historic Times." The situs of this implement, and the age of the tumu lus from "which it was taken, forbid the idea that it could have been modelled after the fashion of a modern dagger. It should be referred to the invention and skill of the ancient peoples who erected the mound and filled the region with specimens of their pro ficiency in flint-chipping.
The latter (Fig. 4, Plate VII.), Professor Jones calls a stone sword. This interesting relic was found by Dr. Jones in an hexagonal stone grave, forming the centre of a Taurial-moimd, located -within an ancient earthwork enclosing thirteen tumuli on the bank of the Big Harpeth liiver, near Franklin, Tennessee. A little less than twenty-two inches in length, this flint implement is an inch and three quarters broad at its -widest part, and is serrated on both edges. It is care fully chipped on either side from the edge toward the central portion, where it is an inch thick. Strong and serviceable was this "weapon. Hafted in horn or wood, it could have been used as a sword; or, attached to the end of a shaft, it would, have constituted a for midable spear. In either case, if properly handled, it would have proved an effective and dangerous weapon. After all, we cannot positively affirm that this serrated implement was not intended to subserve the uses of a saw. In all likelihood, however, it was fashioned to answer the purposes of a lance-head, sword, or dagger.
Fig. 5 of Plate VII. is a typical form of the stone daggers manufactured by the ancient peoples of this semi-tropical region. Made of flint, it closely i*esemt>les some varieties of spear-points and cutting imple ments.
CHAPTER XII.
IGNORAWT of the uses of iron--that most valuable of all metals--the Southern Indians in their agricultu ral, mechanical, and warlike pursuits, were driven to great shifts to supply the deficiency. In this attempt stone, wood, bone, shell, and copper to a limited ex tent, were employed. Implements formed of these materials answered in a rude way the various wants of these primitive peoples, the same tool being often applied to different uses as the necessities of the case and. the poverty of the owner demanded. Of all the ancient incisive implements characteristic of the North American tribes, none is more marked or more gener ally distributed than the STOWE AXE. With the excep tion of arrow and spear points, there is, in the various illustrations furnished by De Bry, a singular ab sence of every thing like stone weapons and. tools. War-clubs, cane knives, hoes made of fish-bones, and wooden paddles are distinctly portrayed, but not a single stone axe is figured. In plate xxvii. (conviviorum apparatus) of the " Brevis Narratio," lying upon
270
ANTIQUITIES OF THE SOTTTHEEN" IHTXEAWS.
the ground near the large pot .upon the fire, is a stout implement like a bill, with, a handle probably two feet in length. Head, point, and handle are all of the same material, and evidently of -wood. The axe figtired in the hands of a native in the act of killing Peter Garnbie in a boat, is certainly metallic, and of European mamifacture. 1
If we relied upon these illustrations for informa tion, they would lead us to believe that stone imple ments were infrequent among the Indians of A'irginia and Florida. Other investigations, the testimony of tumuli and of cultivated fields, and the contents of numerous relic-beds assure us, however, that such was not the fact. Various authorities might be cited con curring in the statement that the manufacture and use of stone axes by the North American Indians were very general. From the many which suggest them selves, we select the following that the historical evi dence on this subject may be fairly presented:
" Instead of Hatchets and Knives," says Father Hennepin,' " they (the Indians) make use of sharp Stones which they fasten in a cleft piece of "Wood with Leather Thongs."
By Loskiel" we are informed that " their hatchets were wedges, made of hard stones, six or eight inches long, sharpened at the edge, and fastened to a wooden handle. They were not used to fell trees, but only to peel them, or to kill their enemies."
In commenting upon the " handicrafts " of the "Vir ginia Indians, Beverly * writes :
MANUFACTURE OF STOINE AXFS.
271
" Before I finish my account of the Indians it will not be amiss to inform you that when the English went first among them they had no sort of Iron or Steel Instruments ; but their Knives were either Sharpen'd JReeds or Shells, and their Axes sharp Stones bound to the end of a Stick, and glued in with Turpen tine. By the help of these they made their Bows of the Locust Tree, an excessive hard Wood when it is dry, but much more easily cut when it is green, of which they always toolc the advantage."
To Lafitau * we are indebted for the following in teresting account: "Stone axes have been in use in America from time immemorial. They are made of a kind of very hard and tough stone, and it requires much labor to mate them fit for use. They are pre pared by the process of grinding on a sandstone and finally assume, at the sacrifi.ce of much time and labor, nearly the shape of our axes, or of a wedge for split ting wood. The life of a savage is often insufficient for accomplishing the work, and hence such an imple ment, however rude and imperfect it may be, is con sidered a precious heirloom for the children. AVhen the stone is finished the difficulty of providing it -with a handle arises. They select a young tree, of which they make a handle, without cutting it. They split one end and insert the stone. The tree grows, tight ens around it, and encloses it so firmly tha,t it hardly can be torn out."
This method, of hafting a stone axe was also prac tised by the Louisiana and Alabama Indians who, according to Captain Bossu," chose a young tree in
Aroeriquains," vol. i., p. 110. Paris, 1V24 (Prof.
272
ANTIQUITIES OF THE SOUTHERN INDIANS.
which--having made an incision with a flint or peb ble as sharp as a razor--they inserted "a stone cut in tlie form of a hatchet." As the tree grew up, it en cased the stone which by that means became insepa rable from it. Afterward they cut off the tree at the proper length, so as to have a handle to the axe of convenient form. The same writer intimates that lance-heads and darts were fastened to their shafts in a similar manner. Du Pratz * describes the axes of the Louisiana Indians as made of a dark-gray stone of flue grain. " Whether these stones," says he, " were naturally flat or were ground on other hard stones, such as the sand-stone found in Louisiana, certain it is they succeeded in malcing axes. These stone axes are an inch or more thick at the head, and half an inch in thickness for three-quarters of their length. The edge is bevelled (forme en bisea'u} but not cut ting, and may be four inches wide, while the head is only three inches in. width. This head is provided with a cavity--deep^ enough to admit a finger--in order to facilitate the fastening of the blade in the split end of the handle; and this end is, moreover, firmly bound, to prevent further splitting. But there is another inconvenience. In using these axes it was not possible to cut wood, but merely to bruise it; and therefore they always hacked the trees close to the ground in order that the fire which, they kindled here might consume more readily the fibres of the wood bruised by the axe. Finally, by dint of labor and patience they succeeded in felling the tree. This labor requires much time; and formerly, therefore, they were much more occupied than at present, being now provided with axes which we trade to them."
1 "Hi3toi.-e de la Lou's'ane," vol. i., F. 166. Paris, 1785.
STONE AXES OF THE CHBKOKEJKS.
273
Writing with special reference to the Chei'okee Indians, Adair* advises us that they " formerly had stone axes, which in form commonly resembled a smith's chisel. Each weighed from one to two or three pounds weight. They were niade of a flinty kind of stone: I have seen several which chanced to escape being buried with their owners, and were carefully preserved by the old people as respectable remains of antiquity. They twisted two or three tough hiccory slips, of about two feet long, round the notched head of the axe ; aud by means of this simple and obvious invention they deadened the trees by cutting through the bark, and burned them, when they either fell by decay or be came thoroughly dry. "With these trees they always kept up their annual holy fire ; aud they reckon it un lawful, and productive of many temporal evils, to ex tinguish even the culinary fire with water. . . . By the aforesaid difficult method of deadening the trees, and clearing the woods, the contented natives got conven ient fields in process of time."
It may be fairly stated that greenstone or diorite was the favorite material used by the Southern Indians in the manufacture of their axes. Tough and durable in its character, this stone best answered the purposes for which implements of the sort we are now" consid ering were designed. Comparatively few were made of flint. Rarely does a chipped axe occur--by far the greater number being ground or rubbed into the de sired shape through the tedious process of attrition with some other stone.
With a view to a more definite description, the stone axes of the Southern Indians may be classified thus:
1 "History of tbc American Indians," etc., p. 403. London, 1775.
274
ANTIQUITIES OF TIIE SOITTIIERK INDIANS.
I. GBOOVED AXES.--These are frequently met with in. the sepulchral tumuli, upon tlie sites of old villages, in relic-teds, and in cultivated fields. In former times they were in very general use. It may be remarked, in passing, that this type, while not unknown, was certainly unusual among the ancient peoples of Europe. In the ninety-six plates -which illustrate the " 1/ake Dwellings of Switzerland, and other parts of Europe, by Dr. Ferdinand Keller," * we seek in vain for an axe of this description. There is a remarkable absence of implements of this kind among the many and interest ing relics so intelligently discussed and presented by ilr. Evans in his recent admirable work upon " The Ancient Stone Implements, "Weapons, and Ornaments of Great Britain." But two are figured by Mr. Nilssoil "--one of cliorite found in the ground, near Gaddarod, in the parish of Horrod, and the other of hornblend, taken from a bog near Lund. He appears somewhat at a loss how to classify them, and inclines to the opinion that they were " wedges with which to split wood." Here, however, no doubt exists in the mind of the observer. The largest grooved axe found within the geographical limits of Georgia, which has passed under the personal observation of the -writer, weighs nearly ten pounds, is ten inches and a quarter in length, six inches wide, and two inches and a half thick. The - groove is an inch and a half wide, and nearly half an inch in depth. The elevated ridges on each side of the groove are three quarters of an inch "wide. In the formation of this groove or transverse furrow, as well as in imparting shape to this implement,
' London, 1866. 3 "Primitive Inhabitants of Scandinavia," plate vjii., figs. 166, 167. London,
GHOOVKD AXES.
275
a pointed flint was used to peck away the portions of the stone sought to be removed. Traces of this pro cess are clearly perceptible, although, after it was com pleted the axe was polished with no little care. This specimen, is represented, by Fig. 1, Plate X., and was taken from a tumulus located at the confluence of the Etowah and the Oostenattla Rivers.
Between this axe and the small but well-formed specimen represented in Fig. 2, Plate X., -weighing only half a pound, the writer has in his collection more than fifty grooved axes, taken from mounds and relic-beds, and picked up in the fields within the pres ent limits of Georgia. Although of different shapes and. -weights they belong to the same class. A fewtypical forms are represented in the accompanying" plate. (See Figs. 3, 4, 5, 6 and 7, Plate X.)
Of axes of this description it may be affirmed that their weights vary from half a pound to nine pounds. Occasionally they will turn the scale even at seventeen pounds. In length they differ from three to twelve or fourteen inches, and in width from two and a half to nine inches. The average width of the groove is about an inch and a quarter; its depth from a quarter to half an inch. The presence of the transverse furrow indicates the manner in which these axes were hafted. If not inserted in the growing tree and there allowed to remain until the wood, had closed tightly around the groove, a strong withe, following the groove, -was bent around the axe, and the ends brought together beneath, where they were firmly lashed by means of deer-sinews or thongs of buckskin. In order to make the implement more secure in the handle, thus formed, it Avill be observed that in many instances the lower or inner side of the axe was carefully
276
ANTIQUITIES OF THE SODTHEIOT ZNDIAJTS.
squared or slightly hollowed out so as to permit tlie insertion of a tightening wedge. (See Figs. 4, 6, and 7, Plate X.) In most oases the groove is near the head of the axe; occasionally, however, this tranverse furrow runs across the central portion, thus affording an opportunity for a double edge. Specimens of this latter variety, so far as our observation extends, are carelessly made, and of soft material. They could liave been used for little else than, offensive pur poses.
"We have already been assured by the testimony of early observers that grooved axes were extensively engaged in deadening forest-trees, so as to clear certain tracts of land for cultivation. They were also em ployed iu removing the bark and bruising the outer fibres near the roots of the trees so that the fires kin dled around them might the more readily eat into the trunks and insure their early fall. It is well known that the Indians of this region, in their mechanical operations, upon every practicable occasion invoked the agency of fire. By its assistance the tree selected for the future canoe was felled, then burnt off at the required length, and finally shaped and hollowed, out. In plate xii. (Lijkterium conficiendorum ratio) of the " Admiranda Narratio," 1 we have a lively representa tion of the entire operation. It is more than prob able that during the progress of such labors these implements, with suitable handles, proved very service able in removing the charred surface from time to time so as to afford fresh fuel for the flame.
An examination of the heads of these axes acquaints us with the circumstsmce that many of them are bruised and splintered, "which indicates that they were used
1 Francoforti ad Mosnum. De Bry, anno 1590.
GBOOVED AXES.----STONE ADZE.
277
either as clubs or as wedges for splitting wood. In the latter case--the edge being placed and by means of the handle held in position--the axe was driven into the wood by blows struck upon its head with a stone or wooden maul, 1
These heads are sometimes rounded, again flat, and at other times wellnigh pointed.
^VTe incline to the belief that the smaller and me dium-sized specimens were tomahawks or battle-axes. Cleverly hafted and at close quarters they would, in stalwart hands, constitute a formidable offensive weap on, whether the blow be delivered from the edge or the head.
Many of these axes are badly worn, thus showing the long-continued use to which they were subjected, and advising us of the fact that their edges were time and again reground or sharpened. So often have some of them been sharpened, that nearly the entire blade has been worn away. The edge was re newed by rubbing it upon, a whetstone. Several of these whetstones are in the writer's collection, deeply furrowed and hollowed by the sharpening of these im plements. The edges of these axes were sharpened evenly on both sides.1
Near akin to the grooved axe is the stone adze, of which Fig. 8, Plate X., may be taken as a fair example. Implements of this sort are rare, and were fashioned in the same manner and of similar material employed in the manufacture of the grooved axes which we have been considering. The specimen before us is made of a tough diorite, is five inches and a quarter in. length,
278
ANTIQUITIES OF THE SOUTHERN INDIANS.
two inches wide, and an met and a quarter in thick ness. It lias been subjected to considerable use, the cutting edge being chipped and -worn, while the lower portion of the flat surface retains the polish derived from long-continued service.
This adze may have been hafted to a wooden han dle, forming with the implement an angle more or less acute at the pleasure of the workman, one end being bent and adjusted to the flat surface opposite the groove, where it was retained in proper position by deer-skin thongs or ligaments of. some sort; or, a flexi ble branch or withe may have been "wound round the groove and the two ends bound together so as tightly to embrace the blade after the fashion generally ob served in mounting the ordinary grooved axe. The bent handle, -we think, was probably adopted.
II. HAND ATTD WEDGE-SHAPED AXES, OB STOXE CELTS.'--In the accompanying Plate XI. are figured six varieties of this class. As in the case of the grooved axes, so with those we propose now to con. aider, greenstone or diorite was the material usually se lected for their manufacture. A few chipped flint axes have been found. The largest specimen represented weighs three pounds and a half, is ten inches and a half in length, and three inches and a half broad at the cut ting edge. Its symmetry of proportion is admirable. Some of these axes are nearly cylindrical, and resem ble very closely the variety called by Mr. Nilsson the cross-axe -with edge ground on both sides. Others have the broad sides somewhat convex, and the nar row sides flat. Some have l^lunt heads and are fanshaped, widening very much at the cutting end.
1 Compare Elans' "Ancient Stone Implements, etc., of Great Britain," chap, vi. London, 1872.
POLISHED STOKE CELTS,
~2 i 9
Others still, terminate in a sharp point at the upper end, as though tlie intention was with it to loosen or "break up the material worked upon, and then, with the cutting end, to remove the particles and smooth the surface. Such a tool would have been very con venient in many instances. Particularly valuable would it have proved for dressing the interior of a wooden canoe hollowed out toy fire. "Within, the old oak canoe, unearthed in 1780, at St. Enoch's croft, and near the prow, lay a beautifully-finished stone axe very similar to the pointed celt we have fig-tired in the accompanying plate. It was doubtless one of the simple implements with, which this primitive Clyde boat had been fashioned. 1 A, like tool was equally effective in giving shape to the cypress canoes which in ancient times navigated the yellow waters of the Savannah and the AAatainaha. These wedge-shaped axes or celts differ in length from three inches to a foot; are, at the cutting end, from two to four inches broad, and in. weight vary from, half a pound to five pounds and upward. The heads are rounded, square, flattened, or pointed. The cutting edge is square, rounded, or semicircular. In all cases, so far as our observation extends, the edge has been ground, from both sides. Occurring frequently in many portions of the Southern States, it is certain that their use was very general among the Indians. The larger and longer varieties were probably managed by hand, and were not hafted. Those of smaller size may have been in serted, in wooden, bone, or horn handles, although even these were entirely capable of manual use without such aid. The absence of a groove and the elongated form are the distinguishing peculiarities of this class.
1 AVUson'g "Prehistoric Man," p. 104. London, 1865.
280
ANTIQUITIES OF THE SOUTHERN INDIANS.
These implements are much tetter adapted to incisive purposes than the grooved axes. Their edges show continued use, and frequent sharpening. Sometimes their heads afford evidence of the fact that they had been struck with a stone or club and thus driven after the fashion of a wedge; but usually, and especially in the case of the longer and larger varieties, the weight of the implement and the strength of the arm sufficed for the accomplishment of the prescribed labor. All of them were first chipped or peeked into shape, and then rendered smooth by the tedious process of attri tion. It is very difficult satisfactorily to discriminate between some varieties of these wedge-shaped axes ' and some forms of stone chisels. So meagre was the supply of tools in the possession of the Indians, and to such various uses were the same implements often and necessarily applied, that it is almost impossible to subject them to a rigid classification.
Professor Joseph Jones discovered in a sepulchral mound on the bank of the Cumberland Kiver, opposite the city of Nashville, Tennessee, an axe of this class with a stone- handle. The entire implement was cut out of a solid piece of greenstone (see Plate XII.). The handle is thirteen inches and a half in length, an inch and a half wide, and about an inch thick. At the lower end is a hole for the suspension and convenient transportation of the weapon when not in actual use. The axe is about six inches long, two inches and a quarter wide at the ciitting edge, and an inch and a half broad at the other end. It is three-quarters of an inch thick, and in general appearance resembles many of the stone celts at one time in such common use among the Southern Indians. This relic possesses spe cial , interest and value, and maybe regarded as per-
PEUFOKATED AXES.
281
petuating t3ie manner in which axes of this class were frequently hafted for domestic and perhaps warlike purposes.
An implement precisely similar in material and construction was taken from a grave-mound in York District, South Carolina, about ten years ago. Kelics of this description are very rare, and 'were fashioned, at the expense of much time and labor. Both of them, were carefully polished in every part. We accept them, not only as curious mementos of a shadowy past, but as enduring proofs of the peculiar mode in which implements of this class were mounted and car ried by these primitive peoples.
It would really appear that the ancient workman, as though mindful of the curiosity which -would-exist in the minds of coming1 generations touching the cus toms and manufactures of an age without letters or established traditions, designed by this permanent legacy to remove all doubt, and bequeath an imper ishable token for the information of those who should come after.
The thin copper axes found in Nacoochee Valley and in a few other localities, are to be referred to the present class. They were inserted in a split handle, and were rather objects of distinction and ornament than serviceable implements. Having, how7ever, in another chapter commented upon these interesting relics at some length, we need here do 110 more than mention their existence.
III. PERFOIIATET> AXES OR HATCHETS. l--It is a noteworthy circumstance that these implements were generally shaped prior to their perforation. It might
1 Compare Evans' "Ancient Stone Implements, etc., of Great Britain," chap, viii. London, 1872.
2S2
ANTIQUITIES OF THE SOUTHEKN INDIANS.
very "well be supposed tliat tlie "workman, anxious to detect any concealed defect in the material which in the end might render useless his entire labor, would have at least blocked out his axe before he entered upon the tedious process of drilling; but that he should not only have fully shaped, but in some instances even polished the "weapon before he commenced drilling the hole for its handle, appears singular. Such, neverthe less, is the fact. As "we write, the physical proofs are before us. Of several specimens now on the desk, one is entirely finished and polished, but lacks the handlehole. A second (Fig. 1, Plate XIII.), pecked into the desired shape, but not yet ground, indicates on the nether side the commencement of the drilling process. Upon a careful examination of a third, it will be per ceived that the drill-hole has been completed only onehalf the required distance. A core or nipple, nearly a quarter of an inch in length, appears at the bottom (Fig. 2, Plate XIII.), clearly showing that a hollow reed, aided by sharp sand arid water, was the instru ment by means of which the perforation was compassed.
Many of these ornamental axes are pick-shaped and made of soft material, such as slate. They vary iu length from three to seven inches. The perforations are made longitudinally through the centre--the points being rounded but not brought to a cutting edge. Fig. 3, Plate XIII., may be taken as a typical repre sentation.
The blades are scarcely more than three-eighths of an inch in thickness; and, in addition to the handlehole, appears a lateral perforation as though for the suspension of the axe. The entire length is rather more than six inches, and the width of the blades an inch and a half. Axes of this shape occur frequently
PERFORATED AXES.
2S3
in the relic-beds along the banks of the rivers where the natives congregated for fishing and hunting. Most of them are broken. Their edges are not sharp. Fash ioned principally of a talcose slate, they were utterly unfit for service and must be regarded as ornamental or ceremonial axes. They vary in size and form, most of them being less than six inches in length and very light. Steatite was also used in the manufacture of these relics.' In a grave-mound in Louisiana, three beautiful specimens of this variety of ornamental or ceremonial hatchets were found several years since. They were made of a ferruginous quartz. Where the two blades united, these implements were reenforced and perforated. There was also a lateral perforation in each blade, at the distance of about three-quarters of an inch from the central perforation. These relics were marvels of symmetry, and polished in the high est degree. Their edges indicated no wear. Evidently these implements, upon whose construction such great care and labor had been bestowed, were not intended for incisive purposes bxit were designed as ornaments or badges of distinction, or for ceremonial uses.
Fig. 4, Plate XIII., closely resembles what Mr. Nilssoii* would call an "Amazon, or two-edged axe." A similarly-shaped implement is represented in the " Sword of Tiberius." Zenophon mentions it in his " Anabasis," and Horace in one of his Odes speaks of ^Amazouia securis. 3 The specimen before us, made of a tough, close-grained diorite, beautifully polished, is four inches long, an inch and three-eighths in diam eter where it is perforated, and an inch and three-
1 See " Ancient Monuments of the Mississippi Valley," pp. 218, 219. ington, 1848.
' Stone Age," p. 71, plato viii., fig. 173. London, 1868, 3 " Carminum," liber iv,, 4. 20,
"Wash
284
ANTIQUITIES OF THE SOUTHERN INDIANS'.
quarters wide at the edges. The handle-hole is fourtenths of an inch in diameter. This axe is stout and strong1, but it will be observed that while the imple ment itself is capable of "withstanding the shock conse quent upon the delivery of a substantial blow, so small is the perforation that no handle, other than one of metal, could prove at all lasting or serviceable. We incline to the belief that this also was an ornamental or ceremonial axe--intended, for display, and not for actual use.
We notice only one other variety (Fig. 5, Plato XIII.), made of syenite, weighing one pound eleven ounces, four inches and a half in length, two inches and three-quarters in width, and an inch and threequarters thick in the middle. The edges of this stout weapon ai'e slightly convex, and five-eighths of an inch thick, the sides of the blades gradually approaching each other from the middle toward the ends. The perforation for the handle is an inch and a quarter in depth, and rather more thau half an inch in diameter. The implement appears to be finished, although it may be questioned whether the maker did not intend, if uninterrupted in his labor, to continue his drilling until the axe was entirely perforated. The bottom of the aperture is concave, showing that a solid drill was used. Circular stride are observable the entire depth of the hole.
After a careful examination of a large number of these perforated axes, we are under the impression that most of them were carried as matters of ceremony, ornament, or distinction; and it may be that the American war-chief suspended from his belt one of these delicate implements, and regarded it with emo tions near akin to those which possessed the breast of
HATCHETS OF EUROPEAN" MANUFACTURE.
285
the Scandinavian warrior as lie cherished and displayed Ms victory-stone.
The use of these stone axes was abandoned very shortly after intercourse was established betwen the red-men and the white traders. Even in Adair's time such implements were rarely to be seen, and those which had escaped interment with their former owners were carefully preserved by the old people and re garded as " respectable remains of antiquity." *
It was the lamentation of the old chieftain at JVlucclasse, that the "white man had not sooner come among the children of the forest to teach them the use of letters, and furnish them with the iron hatchet, the knife, the hoe, and the gun.
Eagerly did the Indians bargain for metallic imple ments ; and the European manufacturers, pandering to the savage taste, fashioned the axes and hatchets in tended for the American market, of those peculiar and often complex types with, which the red-men of the last two centuries have been, in the popular esteem, so inseparably associated. "The warlike arms used by the Cherokees," says Lieutenant Tiniberlake* ("writing in 1761), "are guns, bows and arrows, darts, scalpingknives, and tommahawkes, which are hatchets; the hammer-part of which "being made hollow, and a small hole running from, thence along the shank, terminated by a small brass-tube for the mouth, makes a compleat pipe. There are various ways of making these, accord ing to the country or fancy of the purchaser, being all made by the Europeans; some have a long spear at top, and some different conveniencies on each side. This is one of their most useful pieces of field-furniture,
> " History of the American Indians," p. 405. London, 17V5. 3 " Memoirs," etc., pp. 51, 52. London, 1765.
286
ANTIQUITIES O-F THE SOUTHEKlf INDIANS.
serving all the offices of liatcliet, pipe and sword; neither are the Indians less expert at throwing it than using it near, but will kill at a considerable distance."
CHISELS.--So uncertain is the boundary-line which separates the wedge-shaped axe or stone celt from the chisel, that we are often at a loss to determine the class to which certain specimens should properly be assigned. The truth is, remembering the poverty of their owners and the various expedients to which they were neces sarily compelled to resort in conducting their mechanical operations, we cannot seriously err when "we say that some tools -were used indiscriminately as wedges, handaxes, and chisels. Of the true character and design of some of them, however, "we may speak with at least some degree of confidence.
Numbers 1, 2, 3, and 4, Plate XXV\,maybe regard ed as typical specimens of the ordinary chisels. They are all made of greenstone, carefully polished. Num bers 2 and 4 were, in all likelihood, hafted in sockets of wood, stag's-liorn, or bone, iu like manner as those, of not dissimilar shape, which have been found in the curious and most interesting lake-dwellings of Switzer land and other parts of Europe. Others wanted han dles, and their heads give ample evidence of the fact that they were driven by means of a small wooden or stone maul. These implements are generally thin, varying in length from two and a half to eight inches, and in "width from one to three inches. They are ground from both sides, to form the cutting edge. In. various relic-beds and shell-heaps which I have ex amined (e. g., those on the banks of the Savannah River, especially in Columbia and Richmond. Counties, and on the islands and headlands along the coast). I
CHISELS AKD GOUGES.
2S7
have found the larger bones of the deer, the bear, and the buffalo, fractured longitudinally and split open. The caves of France and Spain afford proof that the bones of animals were there split and crushed by the primitive peoples in order to extract marrow from them. The Laplanders, the Esquimaux, the Austra lians and other savage nations have been doing the same thing ' within the historic period; and it is not improbable that in splitting bones for this purpose, these chisels were in part used by the Southern In dians. Some of these implements are square, with flat sides ; others are cylindrical, with the sides somewhat convex ; others still, being quite thin, are brought to a cutting edge, both at the end, and for a considerable distance on either side. Those made of flint were first chipped and then ground. The greenstone specimens are carefully polished in every part.
The GOUGE differs from the chisel in that it is usually larger and stronger, and by having one side of the lower end scooped out and the other rounded, so as to present a curvilinear edge. I have seen no relics of the Southern Indians resembling the delicate flint hollow chisels described by Mr. Kilsson and other European archseologists. Numbers 5, 6, and 7, Plate XIV., represent the prevailing types. They are generally from four to nine inches in length, and from one to four inches wide. The principal labor in their con struction was expended upon the lower end and in forming a symmetrical edge. "Where the implement was grasped with the hand or hafted, less care was be stowed upon its polish. The upper end almost always has been splintered or broken to a greater or less extent by blows. Some of the smaller specimens may
1 Sh" John Lubblock, " Prc-historic Times," pp. 311, 316, 428. London, 1869.
288
AHTIQTJITIE3 O1T THE SOUTHERN INDIANS.
liavo been hafted, but the larger were evidently in tended for hand-use. Some of these tools were fashioiied with a protuberance or elevation on tlie back-- distant from the cutting edge about a third the length of the implement--by means of which a considerable leverage was gained by simply inserting the lower end in the material to be removed and then pressing down ward with the upper end or handle. (See Fig. 5, Plate XIV.) Diorite was the chief stone from which tliese gouges were made. Comparatively few speci mens occur, and they do not seem to have passed into very general use among the Southern tribes, or at least sucli of them as inhabited the region to which our attention has been chiefly directed.
Bone-gouges are more frequently met with. They are made of the leg-bones of deer and buffaloes (Fig. 8, Plate XIV.).
SCRAPERS.--The spoon-shaped scraper of France and Switzerland is more pronounced in form and purpose than any implement of like character it has been my good fortune to find among the relics. of the Southern tribes." With them, however, scrapers were exten sively used, but commonly in the shape of substantial flint flalces, struck off by a single blow, and with the wider end chipped to a square or rounded cutting edge. "We see also leaf-shaped or triangular implements, thick in the middle--their edges chipped until they were sharp--which were capable of serving the double use of knife and scraper.
The Esquimaux scraper figured by Sir John 1/ubbock, on page 93 of his " Pre-historic Times," is the counterpart of more than one specimen found in the
SCBAPEBS.
289
shell-heaps on the banks of the Savannah River. For the removal of hair from hides, and in sundry ways, such tools "would have proved very serviceable to the primi tive workmen. Shell scrapers were also employed.
Sometimes when a stout arrow or spear head had lost its point, it was repaired and subjected' to a sec ondary use which entitled it to be classed among scrap ers. Fig. 11, Plate XIV., is an illustration of this. At the point of fracture it has been nicely chipped to a cutting edge. In the present instance this edge is semicircular, but the writer has several in his collec tion whose cutting edges are square. This scraper is an inch and a quarter wide, and was made of a beauti ful variegated jasper. Professor Kau has an implement of this sort, which shows most clearly on its edge the polish caused by the continued use to which it had been subjected.
By far the most elaborate scraper I have seen in this region, is that represented by Fig. 14, Plate XIVIt consists of a close-grained dark cliorite, and "was taken from a burial-mound in the Etowah Valley. An implement precisely similar in shape, and some what larger, was unearthed in the same valley, in 1870, near the confluence of the Etowah and Oostenaula ^Rivers. The specimen before us is five inches and a quarter in length, four inches and a half in width, and half an inch thick.
The perforation is nearly half an inch in diameter, and was compassed by drilling from both sides. The cutting edge extends from one shoulder all the way round to the other. The handle is flat and its sides are square. At the nether portion the edge has been much "worn by continual use. The entire implement is well polished.
290
ANTIQUITIES OF TUB SOUTHERN INDIANS.
An implement of similar shape has been represented and classed by Messrs. Squier and Davis among orna mental axes.* With due deference to the opinion of those gentlemen, -we feel constrained to differ from them in this suggestion. Had this been an ornamental axe, suspended for the purposes of display, there -would have been no marked abrasion of the edge. As it is, the proofs of long-continued use are evident all along the lower portion of the edge and for fully two-thirds of the way up, on either hand, toward the shoulders. We incline to the 'belief that it was a scraper, and that the hole drilled through the lower part of the handle was intended to admit the insertion of a buckskin thong by means of which the implement, when grasped, could have been fastened around the -wrist or the back of the hand, and thus the steady and forcible "use of the tool greatly facilitated. Thus employed it would have proved of great value in dressing skins and for diiferent purposes to which a large scraper could have been applied. Figs. 9, 10, 12, and 13, Plate XIV., represent other forms of scrapers manufactured by the primitive peoples of this region.
FLINT KNIVES.--Closely allied to the scraper, and of such construction that they might very readily have been used both as knives and scrapers, are numerous leaf-shaped implements of which Figs. 1, S, 6, 8, and 9, Plate XV., may be regarded as typica]. These are thin, being chipped from the middle toward the sides -where they are brought, all around, to a cutting edge. They vary in length from one to six inches, and in width from half an inch to three inches and a half. Some of them terminate in points so acute that they resemble piercers.
3 "Ancient Monuments of the Mississippi Valley," page 218, Fig. 114, No. 6. Washington, 1848.
T'la.te _ZF.
FLI^T KKTVES.----AWLS.----BORERS.
291
Altliougli the forms of flint knives in use among the primitive peoples of this region are various and often exceedingly rude, numbers 1 to 0, Plate XV\, may be considered the prevailing types.1 The great anxiety of the Indian was to obtain a cutting edge. This secured, he often expended but little labor upon the rest of the implement. Consequently, we meet with many semilunar knives whose backs are thick and square and carelessly chipped. Such were designed to be held in the hand. The backs of others are thin, and these were probably hafted in longitiidinal han dles of bone or -wood. Other knives are almost razorshaped and others still--elongated in form and -with a square cutting edge--required, for convenient use, that the upper end should be inserted in a handle. Some of the larger leaf-shaped implements are so much elongated that it is difficult satisfactorily to determine whether they were intended as spear-heads or as in cisive tools. In plate xv. of the " Brevis ^arratio," De Bry furnishes a frightful illustration of the enor mities perpetrated upon the bodies of their slain ene mies by the Florida Indians by means of arrows, clubs, and cane knives (arundinis frag'mentisy, but he no where, so far as we now remember, figures a single flint implement which could be called a knife.
A.WLS, OR BORERS.--~V\7e are informed in the early Spanish narratives that the Southern Indians, with heated copper spindles, perforated pearls so that they might be strung and worn as ornaments. That this was not the only kind of piercing implement fashioned by the natives, is evidenced by the presence of flint awls or borers, four forms of which are here figured
] Compare Elans' "Ancient Stone Monuments, etc., of Great Britain," chap ter xv. London, 1872.
292
ANTIQUITIES OF THE SOUTHERN
(see Figs. 2, 3, 4, and 5, Plate XVI.). Number 5 may have answered the double purpose of awl and scraper. 1 The point, as well as the square edge at the opposite end, exhibits that peculiar polish which is born only of prolonged use and attrition.
Ordinary piercing implements were also formed from bone, and of these number 1, Plate XVI., is an excellent example. It is made of a deer's tibia, and is seven inches and a half in length. The scars left upon its surface by the flint implement employed in shaping and polishing it, are still very perceptible. Sharppointed fish-bones were also extensively used, and these are often found in the shell-heaps and relic-beds both on the coast and along the banks of fresh-water rivers. Flint saws are not infrequent.
Before concluding this brief notice of the cutting and piercing implements of the Southern Indians, it is proper to notice a class of tools--similar in general features to the ordinary hand-axes--riiiade sometimes of slate, at other times of a hornblendic stone, again of diorite, and rarely of flint, whose edges are blunt and rounded, or square. They were, to all appear ances, designed as smoothing or polishing stones (see Figs. 6, 7, and 8, Plate XVI.), and may have been used in dressing skins. Their edges are all worn very smooth from constant attrition. The implement, represented in Fig. 9, Plate XVI., typifies a large class, examples of which abound in the relic-beds on the Savannah River. Their use is not well ascertained, but their flat surfaces are very smooth as though they had been constantly employed in rubbing. There are
1 The similarity between this implement and tliat figured by Mr. Evans on rage 289 of his "Ancient Stone Implements, etc., of Great Britain" (London, 1872), is very striking.
XW.
DRIFT IMPLEMENTS.
293
also stout triangular-shaped flint articles, which may be regarded as primitive axes, as unfinished spear-heads, or as scrapers. This matter of classification is, to a considerable extent, arbitrary; and while in most in stances we have no hesitancy in determining the uses and characters of various relics, we not infrequently encounter specimens concerning whose specific employ ment and accurate archseological arrangement any thing more than a suggestion appears unjustifiable.
The implements we have been examining, were ob tained from mounds, shell-heaps and relic-beds, gath ered upon the sites of ancient villages and fishingresorts, or ploughed up in cultivated fields. Before bringing the present chapter to a close, -we desire to allude to some rudely-chipped, triangular-shaped imple ments found in ^Nacoochee Valley under circumstances which seemingly assign to them a very remote anti quity. In material, manner of construction, and in general appearance, so nearly do they resemble some of the rough, so-called flint hatchets belonging to the drift type, as described by M. Boucher de Porthes, that they might very readily be mistaken the one for the other.
Through this valley flows the Chattahoochee. The region being auriferous, the attention of the early set tlers was soon attracted to an examination of the bed of this stream. Particles of gold -were found inteimixed with the sand and pebbles which lay at the bottom. In order to facilitate mining operations, canals were cut, sometimes deflecting the current from its channel, and at other times branching off from the river so as to unearth the precious metal -which had gravitated out of sight. These sections passed through the soil and the underlying drift composed of sand,
294
ANTIQUITIES OF THE SOUTHIOEN INDIANS.
grave], and bowlders, and reached down to the hard slate-rock below. During one of tliesc excavations, at a depth of some nine feet below the surface, inter mingled with the gravel and bowlders of the drift and just above the rocky substratum upon which the de posit rested, were found three flint implements, simi lar in shape, one of which is here figured (Fig. 10, Plate XVI.). It is three inches and a quarter in length, and two inches and an eighth in "width. It is said that articles of a like character have been discovered in the drift along the line of Duke's Creek, but they have not passed under the writer's observation. In this drift, so far as my knowledge extends, no human bones have as yet been found. Prominent earth-mounds, stone graves, and frequent relics attest the fact that this valley was for a long period thickly populated by the red race. These indications of a former occupancy are chiefly confined, however, to the surface, or its vicinity. "When the white men possessed themselves of this beautiful region, these mounds were covered with trees, to all appearances as vigorous and as old as those which composed the adjacent forests, Indian inhumations outside of the tumuli are shallow. Spear and arrow heads, stone mortars, pipes, beads, discoidal stones, axes, and various relics indicating use and orna ment, are confined to the mounds, graves, and the sur face of the valley. Such do not obtain, in the drift. I am not in possession of data sufficient to warrant the expression of an opinion touching the age of Nacoochee Valley as at present constituted. That it has under gone no material change for centuries, is demonstrated by the presence of these large earth-mounds and the big forest-trees which grew upon them after they were neglected or abandoned by those who erected them.
DBII-T iMpuearENTs.
29j
The Chattalioochee lias been pursuing its present course through this charming valley for lo ! these many, many years, and. there are no indications of any violent and sudden mutations which would have modified the period requisite for the gradual formation of the soil and sur face of this valley. Xhat tlie implements in question were brought down with and deposited in the drift when as yet there was little or no vegetable life in the valley, seems highly probable. How many centuries have looted, clown upon the gradual accumulation of the soil which now overlies the drift, none can answer; but of one thing we may rest satisfied, that these specimens of the rude labor of prehistoric man may well claim high antiquity. They are as emphatically driffc inplements as any which have appeared in the diluvial matrix of France. Thus, in Nacoochee, wlrile the Neolithic age is richly represented, the iPalscolitliic period is not entirely wanting in its characteristic types.
If we are ignorant of the time when the Cliattahoochee first sought a highway to the Gulf--if we know not the age of the artificial tumuli which still grace its banks--if we are uncertain when the red nomads who in. fear and wonder carried the burdens of the adventurous De Soto as he conducted his followers through primeval forests and by the side of this softlymoving stream, first became dwellers here--how shall we answer when questioned as to the age in which these rude drift implements were fashioned and used by the primitive peoples ?
CHAPTER XIII.
Agriculture and Agricultural Implements.--Ceremony of the Busk.--Cultivation of Maize.--Mortars and resiles.--Crushing-Stones.--Snt-Stoucs.--Use of Walnut and llickory-iml Oil.
TJGH in the mythology of tlie red-men of tlie South tlie "beneficent Ceres, who first taught mortals how to turn the soil with a plough,1 received no indi vidual deification, they were not insensible of her be nignant influences, ever present in the genial warmth of bright sties,' engendering fertility in the soft earth and causing to spring up beneath their feet a beauti ful plant whose fruit proved in very deed a " stair* of life." By nothing was the gradual development of the semi-civilization of the ]VTuscogulges, the Creeks, the Choctaws, the Cherokees, and other Indian tribes more clearly indicated than by their general and regular cultivation of the maize, an American plant, whose value--recognized by these aborigines for many ante cedent centuries and extensively appreciated at the clawn of the historic period--has ever since received ready acknowledgment wherever introduced to the notice of civilized man. Regarded as a direct gift from the Author of Life to his red children, it was highly
1 " Publii Virgilii Mamma Georgica," lib. i., v. 146. London!, spud A. Dulau & Co., 1SOO.
CULTIVATION OF MAIZE.
297
prized and held in peculiar esteem. To make light of, or waste either the grain, or the cob from which it was taken, was never permitted. Certain ceremonies were observed in the spring when it was planted; and of all their rites the Busk--celebrated just be fore they garnered the ripe ears from the fields--was, perhaps, the most solemn and imposing. Of the American Indians the Southern nations were the most civilized and the least nomadic in their habits. En joying a mild climate, and possessing fruitful and wellwatered valleys, they located permanent seats, were provident of the futnre, and surrounded themselves with more of the comforts and conveniences of life than appertained to the Northern and "Western hunter tribes. Attached to the soil, often building considerable to\vns fortified by palisades, and composed of huts and houses substantial after their kind, and furnished with mats, benches, and various aptly-made domestic utensils, they lifted themselves at least somewhat above that rude, beggarly, and precarious existence which so painfully characterized the condition of so many of the aborigi nes inhabiting other portions of this country, oppressed by greater penury and contending against the rigors of more tempestuous seasons.
The territory over which cultivation by the natives extended, is bounded on the east by the Atlantic, on the south by the Grulf of Mexico, on the west, gen erally by the Mississippi, or perhaps more properly by the prairies, and on the north by the nature of the climate.1 The population of those regions in which the , soil was cultivated, was more permanent and numer ous than that of localities where the individuals relied
* See Mr. SoUatin's "Synopsis of the Indian Tribes," " Ard.ajologia Ameri cana," vol. ii., p. 149.
29S
ANTIQUITIES OJ1 THIE SOUTHERN INDIANS.
for tlieir subsistence upon the natural products of the earth and the waters. It -was also less "barbarous. The most cursory examination of the early accounts will advise us of the fact that maize was extensively cultivated, by, and formed a standard article of food among the Southern Indians. The English at James town were, at times, almost "wholly sustained by the liberality of the natives; and Captain John Smith, in recounting the friendship of Pocahontas, mentions the circumstance that she in person accompanied from the Indian fields the " conductas" of grain which, relieved the wants of the colonists. Both Cabeea de Vaea ' and Captain Eibault" found it growing freely in Florida. From Tampa Bay, De Soto addressed a letter to the Justice and Board of Magistrates in Santiago cle Cuba, informing them that Baltazar de Grallegos, whom, with eighty lancers and a hundred foot-soldiers, he sent to reconnoitre the country, had seen " fields of maize, beans, and pumpkins, with other fruits and pro visions, in such quantity as would suffice to subsist a very large army without its knowing a want." 3 On one occasion his army marched for two leagues through continuous fields of corn. During the progress of the expedition the Spanish soldiers subsisted almost ex clusively upon food furnished by the natives. The maize, stored in granaries and standing in cultivated fields, furnished bread for the troops, while the blades of the corn proved excellent forage for the horses.
"While passing through the pine-barren regions, where the soil was poor and the population scant, bitterly did the Christians complain of the hardships .
CULTIVATION OF MAIZE,
299
there encountered. Often were they sorely pressed to escape starvation. Ample fields and houses filled, with corn were frequently met farther on, and there heavy were the contributions levied, and. numerous the cap tives made who "were compelled, even in chains, to accompany the conquerors and bear weighty burdens of maize and mortars in which to prepare it for cook ing. It would appeal' from the early narratives that the principal towns and maize-fields of the natives were located in rich valleys where a generous, soil yielded with least labor the most remunerative har vest. AVhile beans, pumpl^ins, dried plums, grapes, persimmons, mulberries, nuts, and other spontaneous products of the earth, were freely used, it is quite cer tain that the Southern Indians relied chiefly upon their crops of corn. Upon its cultivation general and sys- tematic attention was bestowed. In a former chapter we have seen that the grooved axe was extensively employed in girdling trees so as to deprive them of life, and thus, in the end, cause the forest-growth to disappear from the spots which had been selected for cultivation. Indian fields in which not even the trace of a stump or root could be perceived were frequently observed by the first European settlers. For the loca tion of such fields the richest spots adjacent to the villages were selected. These were planted in com mon--no fences,' in the olden time, indicating the bounds of individual labor, or private storehouses the fruits of personal toil. The soil was the property of all--and each, sharing in the general toil, partici pated in the common harvest. " A.bout their houses," says Captain Ribault, "they labor and till the ground, sowing their fields with a grain called J^fahis, whereof
1 SeBrii;kell's " Natural History of North Carolina," p. 344. Dublin, 1M7.
300
ANTIQUITIES OF THE SOUTHERN IjSTJJlANS.
they make their nieal, and in tlieir gardens they plant beans, gourds, cucumbers, citrons, peas, and. many other fruits and roots unknown to ns. Their spades and mattocks be made of wood, so well and fitly as is pos sible, which they make with certain stones, oystershells and muscles, wherewith also they make their bows and small lances, and cut and polish all sorts of wood that they employ about their buildings and ne cessary use." 1
The Gentleman of Elvas a intimates that each In dian had his own field which he planted and harvested for his individual account. The natural fruits, he con tinues, were common for all. In some parts of the territory traversed by Cabec,a de V^aca 3 three crops of maize and beans were raised during the year. A Natchez chief, among other things, offered jVI. Le Page Du Pratz 4 twenty barrels of maize in exchange for a sun-glass.
W^e are informed by .Adair E that while the gar dens contiguous to the houses were fenced in, the large fields were, in this regard, quite unprotected. In plate xxi. of the " Brevis !N"arratio," six Indians are seen busily engaged in preparing the ground and in plant ing corn. No fences or enclosures of any sort are rep resented. It would appear from the explanatory note G
1 "The "Whole and True Discoverye of Terra Florida." London, 1863. '> "Narratives of the Career of Eernando de Soto," translated by Bucking ham Smith, p. 201. New York, 1866. 3 " Relation," etc., translated by Buckingham Smith, p. 172. New York, 1871. 4 "History of Louisiana," vol. ii., p. 183. London, 1763. 5 " History of the American Indians," p. 406. London, 1775. a " Diligentcr cohint terrain Indi, earn ob causam Hgcmes o pisciam ossibas parare norunt viri, quibus maimbria lignea aptantes, terrain fodiunt satis facile, natn
Mayzum scrunt, preeeuntibus iionnullis, quje defixo in terram baculo foramina faciuntj inqnse fabje & milij grana inijciantur. Fact a sementc, agros relin quunt," etc.
AGBICULTTJBAL IMPLEMENTS.
801
that the Indians diligently cultivated the soil, using for this purpose nsh-bones attached to wooden handles. With these agricultural implements the men broke up and made even the surface of the ground. Following after them came the women who, with the aid of sticks, made holes in the newly-prepared and soft earth, into which beans and grains of corn, carried for that purpose in small baskets, were dropped. The plant ing being over, the seed was left to fructify--but little attention being bestowed upon the growing crop.'
K~o specimens of these bone agricultural, imple ments or of the wooden spades and mattocks ' men tioned by Captain Uibault have passed under our ob servation. These, as well as the scapulas of the deer and the buffalo, which were used for a similar purpose, have crumbled into dust. Occasionally, however, we meet with stone hoes, of which Kg. 1, Plate XVII., may be regarded as typical. This 1'elic is made of greenstone. It is five inches and a quarter in length, and nearly two inches arid three-quarters in width. For a distance of more than two inches and a half from the edge it exhibits on both sides that delicate polish which is engendered only by constant attrition and long-continued use. The groove afforded the means of lashing it securely to a handle \vhose end was doubtless bent for that purpose, so that the blade should remain at right angles to it. It will be ob served that this implement is slightly curved, and has very much the appearance of the half of a grooved axe split in twain longitudinally. It is, nevertheless,
> Compare " A Briefe and True Report of the Xew-found Land of Virginia," etc., "made in English by Thomas Harlot," ete., pp. 14, 15. IVaneoforti ad Mojnnm. J)e Dry, anno 1590.
* See also Brickell's " Natural History of Xorth Carolina," p. 826. Dublin, 1T8V. Bossu's " Travels through Louisiana," etc., vol. i., p. 224. London, 1771. Loskiel's "History," etc., p. 68. London, 1704.
302
ANTIQUITIES OF THE SOTJTHEBJST INDIANS.
a complete and. well-formed, lioe. Remembering tlte shallow manner in widen tlie natives cultivated, the soil, we can readily believe that it would have abun dantly answered tlie purpose for which -we suppose it to have been designed. Fig. 2, Plate XVII., represents a spade made of greenstone which was found by Prof. Joseph Jones in a Tennessee grave-mound. "Were the handle shorter, it might be classed as a scraper or smoothing-stone. This implement is beautifully pol ished. Its entire length is seventeen inches and a quarter, the handle--which is round and tapering--be ing fourteen inches "and a quarter long, and the blade three inches long and nearly as wide. The blade, both on its sides and bottom, was brought to an edge. ~We suppose this to have been an agricultural tool.
Large and roughly-chipped, leaf-shaped flint imple ments, of which Figs. 3, 4, and 5, Plate XVII.,,may be taken as types, are found in considerable numbers. These, we think, should be properly classed among primitive agricultural tools. None of them, however, so far as my observation extends, are as well formed or clearly marked as the notched implements from East St. Louis, so well described by Prof. Kau in the Smithsoniau Report for 1868.'
After favoring us -with an . account of tne manner in which the Louisiana Indians constructed their huts, Du Pratz" says: " ]STear all their habitations they have fields of maiz and of another nourishing grain called Olloupicho^lT^ which grows without culture. For dress ing their fields, they invented, houghs which are formed, in the shape of an L, having the lower part flat and sharp; and to take the husk from their corn they made
'P. 401, ttttq. s "History of Louisiana," etc., vol. ii., p. 225. London, 1763..-
CEREMONIES OF THE BOOS--KE--TATT.
303
large wooden mortars, by hollo-wing the trunks of trees with fire."
The corn having attained its maturity, and being ready for harvest, a day was named Try the mico for the celebration of the annual festival known among the Creeks as Boos-ke-tau. In Cussetuh, eight days were spent in conducting the prescribed ceremonies, while in towns of lesser importance four days sufficed for the observance of this memorable season of purification, thanksgiving, and rejoicing.
On the morning of the first day,1 the warriors clean the yard of the square and sprinkle it -with white sandThe a-cee, or decoction of the cassine yupon, is made. The fire-maker kindles the fire, as early as he can, by friction. Four logs, each as long as a man can cover by extending his two arms, are cut and brought by the warriors and placed in the centre of the square, end to end, thus forming a cross. The outer ends indicate the cardinal points. In the centre of the cross the new fire is made. These four logs are burnt out during the first four days.
The Pin-e-bun-gau (turkey-dance) is danced by the women of the turkey tribe, and while they are dancing the possau is brewed. This is a powerful emetic. From twelve o'clock to the middle of the afternoon the possau is drunk. After this four men and four women dance the Toc-co-yule-gau (tadpole). From evening until daylight E-iie-hou-bun-gau (the dance t>f the peo ple second in command) is danced by the men.
About ten o'clock, the second day, the women dance Its-ho-bun-gaTi (the gun-dance). After twelve, the men go to the new fire, take some of the ashes, rub
304
ANTIQUITIES OF THE SOUTHERN INDIAN?.
them on the chin, neck, and belly, jump head-foremost into the river, and then return into the square. The women having prepared the new corn for the feast, the men take some of it and rub it between their hands and then on their faces and breasts, and then they feast.
During the third day the men sit in the square. Early in the morning of the fourth day the women get the new fire, clean out their hearths, sprinkle them with sand, and kindle their fires. The men finish burning out the first four logs, and then rubbing them selves with the ashes on their chins, necks, and bellies, go into the w~ater. This day salt is eaten, and they dance Obuugauchapco (the long dance). The fifth day four new logs are brought and placed in the same position as on the first. They drink also a-cee, the strong decoction, of the cassine yupon. During the sixth day they remain in the sqnare. The seventh day is passed in like manner. On the eighth day they get two large pots and their physic-plants, to wit: Mic-co-ho-yon-e-juh, Toloh, ^-che-nau, Cup-pan-pos-cau, Chn-lis-sau, Tuck-thlau-luste, Tote-cul-hil-lis-so-wau, Chofeinsuck-cau-fuck-au, Chofe-nius-see, rlil-]is-hut-ke, To-te-cuk-chooc-his-see, A^elaunuh, Oak-chon-utch-co, and Co-hal-le-wau-gee. These are all put into the pots and beaten up with water. The chemists (E-lic-chul-gee, called by the traders physicmakers) blow into the decoction through a small reed, and then the men drink it and rub it over their joints until the afternoon. They then collect old corn-cobs and pine-burs, and, placing them in a pot, burn them to ashes. Four virgins who have never had their men ses bring ashes from their houses, and, having put them into the pot, stir all together. The men take "white clay, and mix it with wrater in two pans. -A. pan
CKTCEMOiNIES OF THE BOOS--KB--TATT.
305
of this clay and one of ashes are can-led to the cabin of the mico. Two pans similarly filled are taken to the cabin of the warriors. ^Vith the clay and ashes they rub themselves. Two men, appointed to that office. Taring1 flowers of tobacco of a small kind (Itchau-chu-le-puc-pug-gee) or, as the name imports, the old man's tobacco, which -was prepared on the first day, and putting it in a pan on the niico's cabin, give a little of it to all who are present.
The mico and councillors then go four times around the fire, and every time they face the east. throw some of the flowers into the fire. They then go and stand to the w'est. The same ceremony is re peated by the warriors.
A cane is stuck up at the cabin of the mico, with two white feathers in its end. A member of the Fish tribe (Thlot-lo-ul-gee) takes it just as the sun goes down and moves off toward the river, all following him. "When half-way to the river, he gives the deathwhoop. This he repeats four times between the square and the "water's edge. Here, they all locate themselves as close together as they can stand. The cane is stuck up at the water's edge, and they all put a grain of the old man's tobacco on their heads and in each ear. At a given signal, four times repeated, they throw some of this tobacco into the river, and every man upon a like signal plunges into the stream and pick's up four stones from the bottom. With these they cross themselves four times on the breast, each time throwing a stone into the river and giving the death-whoop. They then wash themselves, take up the cane with the feathers, return and stick it up in the square, and visit through the town.- At night they dance O-bun-gau-Haujo (the mad dance), and this
306
ANTIQUITIES. OF TIIJ5 SOTJTHEBj^ INDIANS,
finishes tlie ceremony. This happy institution of the Boos-ke-tuh' restores man to himself, his family, and to his nation. It is a general amnesty which not only absolves the Indians from all crimes, murder excepted, but seems to bury guilt itself in oblivion. In ancient times this festival was celebrated at the appearance of the first new moon during which, the corn became fully eared. Subsequently, however, it was regulated by the season of the harvest.2 From the time con sumed and the formalities observed in its solemniza tion, it is manifest how important and sacred this ITeast of the Buslc/vvas in. the estimation of the agricul tural tribes of the South. The ingathering of the matured maize-crop was preceded by an extinguish ment of former fires and the kindling of one consecrated new fiame, which was to prove the parent of light and heat for the coming year. This "was the season of physical and moral purification, of general forgiveness, universal amnesty and united thanksgiving. Then \vas the blotted chapter of the old year closed and sealed, and a new, clean page opened in the life of every one. !N"o wonder that these primitive peoples held this maize in special honor and "watched its growth with emotions other than and superior to those which would have been suggested, had they regarded it
1 For other accounts of the solemnization of this festival, see Du Pratz' "His tory of Louisiana," vol. ii., p. 180, et scq. ; Schoolcraft's "Archives of Aboyginal Knowledge," vol. v., p. 267, et seq. ; Adair's "History of the American Indians," p. 99, etscq.; Bartram's "Indians of the South," part 1, of vol. iii. of the Transac tions of the American Ethnological Society. Bartram's " Travels through North and South Carolina, Georgia,1 ' etc., etc., p. 507, et seq. ; Briekell's " Natural His tory of North Carolina," p. 326.
For the ceremony and preparation of the Black-Drink, see Schoolcraft's "Ar chives of Aboriginal Knowledge," vol. v., p. 266, et seq. ; " Brevis Narratio," plate xxix.
2 Adair'd " History of the American Indians," etc., p. 99. London, 1*775.
HARVESTING THE CEOPS.----PUBLIC GRANABIES. 307
simply as an ordinary plant and a common article of food.
This festival over, immediate attention was directed to harvesting the crop. Bartram * says that the whole town then assembled, and every man carried to his own granary the fruits of his labor from the part of the general plantation allotted to him in the spring. This share of the harvest became his individual prop erty. Previous, however, to their carrying off their crops from the field, he continues, " there is a large crib or granary, erected in the plantation, which is called the king's crib ; and to this each family carries and deposits a certain quantity according to his ability or inclination, or none at all if he so chooses : this in appearance seems a tribute or revenue to the Mico ; but, in fact, is designed for another purpose, i. e., that of a public treasury supplied by a few and voluntary contributions, and to which every citizen has the right of free and equal access when, his own private stores are consumed ; to serve as a surplus to fly to for suc cour ; to assist neighbouring towns, "whose crops may have failed ; accommodate strangers or travellers ; afford provisions or supplies when they go forth on hostile expeditions; and for all other exigencies of the state ; and this treasure is at the disposal of the king or Mico."
It is probable that this harvest-labor formerly de volved to a large extent upon the women. In plate xxiii. of the " Bre vis Narratio," women, and men of that bestial class improperly styled in the early narratives Herma/plirodites, are busily engaged in the transporta tion of baskets filled with fruits. The preceding plate exhibits to us a storehouse---located on the low bank of
1 "Travels," etc., p. 510. London, 1792.
308
AiNTIQTJITIES OF THE SOirTHEEW
a stream--toward which several canoes filled with fruits and corn are tending. These granaries or storehouses among the Florida Indians were built of stones and earth, and covered with palmetto-leaves and clay. For their erection some cool spot was selected where pro tection was afforded, against the violent rays of the sun. Such storehouses served as depositories not only for maize, fruits, nuts, and roots, but also for dried fishes, alligators, dogs, deer, and other jerked meats. These were first exposed upon a scaffolding,' made of poles, beneath which a fire was kindled and kept burning until the meat, thoroughly smoked and dried, was thus preserved from early decomposition.
These hoards of corn, meat, arid fruits, are frequent ly mentioned in the early narratives. In the language of the " Gentleman of Elvas," ' " four leagues before coming to Chiaha fifteen men met the Governor--bear ing loads of maize--with word from the cacique that he waited for him, having twenty barbacoas full." Garcilasso de la Vega 3 states that one of De Soto's officers found in one house five hundred measures of ground maize, besides a large quantity in the grain. Law-son * says that the cabins intended for granaries ' were made without windows.
The maize thus constituting a chief source of sub sistence among the Southern Indians, it is interesting to note the method generally adopted by them in pre paring it for cooking. The Fidalgo of Elvas makes the broad assertion that the bread which is eaten
" Brevis Narratio." of the Career of nernanjo de Soto, 1 ' translated by Buckingham
1781. 1866. L," p. 3'2~. Dublin,
PREPARATION OS MAIZE FOR FOOD.
309
throughout Florida is made of maize ; and, at Apalachen, Cabeca de Vaea observed numerous mortars for cracking this grain. In plate xxviii. of the " Brevis Narratio" (con'vivioru'm apparatus), a flat, round stone mortar, set upon tie ground, is represented among other articles. A native, 011 bended knee, with a short, stout pestle in his hand, is in the act of grinding something for the feast. The intimation is, however, that he is at present simply "bruising * some fragrant herbs to serve as a seasoning for the food which is boiling in the great clay pot. Among the North and South Cjirolma Indians " the savage men never beat their corn to make bread, but that is the women's work, especially the girls, of whom you shall see four "beating with long, great pestils in a narrow wooden mortar; and every one keeps her stroke so ex actly, that 'tis worthy of admiration."
" Their common food," says Captain Bernard Ro mans,3 " is the 2ea or Indian corn, of which they make meal and boil it; they also parch it and then pound it; thus taking it on their journey they mix it with cold water, and will travel a great way without any other food ; . . . they have also a way of drying and pound ing their corn before it comes to maturity ; this they call Itoota copassa (i. e., cold flour) ; this in small q^xiantities thrown into cold water boils and swells as much as common meal boiled over a fire; it is hearty food, and being sweet, they are fond of it," etc.
To Adair* are we indebted for the following ac count of the mortars in which the women beat the
London, ma.
310
ANTIQUITIES OF THE SOUTHERN" INDIANS.
flinty corn until all the liusks -were carefully taken off, and the cracked grains--well sifted and fanned-- neatly prepared for boiling in. large earthen pots : " The Indians always used mortars instead of mills, and they had them, with almost every other conven ience, when we first opened a trade with them ; they cautiously burned a large log to a proper level and length, placed fire a-top, and wet mortar round it, in order to give the utensil a propel' form; and "when the iire "was extinguished, or occasion, required, they chopped the inside with their stone-instruments, patiently con tinuing the slow process till they finished, the ma chine to the intended purpose."
Of this maize, the same writer informs us, the In dians of Upper Georgia and the adjacent region pos sessed throe varieties : the first was small, matured in two months, and was called by the English " six weeks' corn;" the second was yellow and flinty, and known among the natives as " hommony-corn;" "while the third, "which was largest and yielded a white, soft grain, was called " bread-corn." " In July, when the chestnuts and corn are green and full-grown, they halfboil the former and take oft' the rind; and having sliced the milky, swelled, long rows of the latter, the women pound it ill a large wooden mortar, which is wide at the mouth and gradually narrows to the bot tom ; then they knead both together; wrap them up in green corn-blades of various sizes, about an inch thick, and boil them well, as they do every kind of seethed food. This sort of bread is very tempting to the taste, and reckoned most delicious to their strong palates. They have another sort of boiled bread, which is mixed with beans or potatoes ; they put on the soft corn till it begins to boil, and pound it sufficiently fine. . . .
MORTARS AND PESTLES.
311
the flour is stirred and dried by the lieat of the sun or fire, they sift it with sieves of different sizes curiously made of the coarser or finer cane splin ters. The thin cakes mixt with dear's oil were former ly baked on thin broad stones placed over a fire, or on broad earthen bottoms fit for such a use. . . . When they intend to bake great loaves they make a strong, blazing fire, with short, dry, split wood on the hearth. When it is burnt down to coals they carefully take them off to each side and sweep away the remaining ashes; then they put their well-kneaded broad loaf, first steeped in hot water, over the hearth, and an earthen bason above it, with the coals and embers atop. This method of baking is as clean and effica cious as could possibly be done in any oven; when they take it off they wash the loaf with warm water, and it-soon becomes firm and very white. It is like wise very wholesome and well-tasted to any except the vitiated palate of an Epicure."
While it is well ascertained that -wooden mortars and pestles were in general use among the Indians I at the period of our first acquaintance with them, and furnished a ready method for husking and pounding their maize, it is equally certain that at some remote time mortars, pestles, and crushing implements, made of stone, were not uncommon. Dr. Dickeson and other explorers have found them in the tumuli of Alabama, Mississippi, Louisiana, and Texas. They have been taken from the mounds of South Carolina and Florida. From a single relic-bed on the right bank of the Savan nah Eiver, a few miles above Augusta, I obtained, at
OJ-2
ANTIQUITIES OF THE SOUTHERN INDIANS.
oae time, thirteen stone mortars made of flat "bowlders taken from the bed of the stream and hollowed out on both sides to the deptli of two or three inches. (See Figs. 1 and 2, Plate XVIII.) The average diametar of these shallow, basin-like excavations was rather more than nine inches. No labor had been expended in shaping the stones. The natives took them'as they fjund them, and simply formed the cavities. Placed npon the ground or held in the lap, with the assistance of the ordinary disk-shaped crushing-stones -- large numbers of which were seen in the vicinity--the green corn could have been mashed, the parched corn pounded, or the husks beaten from the ripe grains. This rude variety is frequently met with in many por tions of the State. Belonging to the same class, ex cept that it has been hollowed out only on one side, is the mortar represented in Fig. 3, Plate XVIII.
The bowl is scarely more than an inch in depth, and about five inches in diameter. By far the most symmetrical and carefully-fashioned mortar I have seen was ploughed up in a field in Liberty County, some ten miles from the sea-coast. IVtade of a yellow, ferru ginous quartz, with a flat bottom and circular walls gradually expanding as they rose, its general shape was that of an inverted, truncated cone. Entirely arti ficial, the exterior was well polished. About ten inchea" high, eight inches in diameter at the top, and seven inches at the bottom, the interior had been ex cavated to the depth of nearly eight inches. At the top the walls were about three-quarters of an inch thick, and increased in thickness as they descended. No material exists in this section of the State from which such a utensil could have been manufactured. The probability is, that it was made at a considerable re-
MOKTABS.
313
more from the spot where it was found, and was sub sequently brought to the coast by some primitive mer chantman, by whom it was there exchanged for seashells and other articles of value native to this region. Upon its construction, great time and labor must have been expended ; and this relic is a remarkable illus tration of the still and patient industry of the ancient workman, who, unassisted by any metallic tools, was able from such a hard sxibstance to fashion a mortar so serviceable and so admirable in all its proportions. A mortar not unlike in its general appearance was ob tained by the Kev. George Howe, D. D., from an In dian cemetery on the bank of the Congaree Kiver, a few miles from Columbia, South Carolina. It is figured in the plate facing page 178 of the sixth volume of Mr. Schoolcraft's " Archives of Aboriginal Knowledge." In a subsequent chapter we will observe that some of the imperforate discoidal stones clearly indicate that at some time or other they have been diverted from the original purpose for which they were manufac tured, and have been, treated as mortars. This sec ondary use entitles. such to specific mention in. this connection. In addition to the stone mortars de scribed, I have seen, in the middle and upper parts of the State, large bowlders--some of them waist-high-- permanent in their location, whose tops had been hollowed, out for mortars. These cavities "were circular in form, and capable of holding a half-peck or more. They may be regarde'd as public property, and. afford proof of the stability of the agricultural population by which they were used. Hunter * alludes to the presence of wooden mortars among the tribes west of the
* " Manners and Customs of Several Indian Tribes located west of the Missis sippi," pp. 269, 270. Philadelphia, 1823.
314;
ANTIQUITIES OF THE SOUTHERN" INDIANS.
Mississippi, and says that, "in addition, each village lias one or two large stone mortars for pounding corn ; they are placed in a central situation, are public prop erty, and are used in rotation by the different families." Mr. Bartlett counted, at El Paso, twenty-six artificial cavities in detached blocks of stone which had been hollowed out by the Indians, and served as mortars in which to pound their maize.' The pestles handled in connection with the wooden mortars, consisted of pieces of hard wood, between three and four feet long, heavy and rounded at each end, and narrow in the middle where they were grasped.
Stone pestles--of which Figures 4 and 5, Plate XVIII., are typical representations--were both shorter and narrower, varying in length from seven to eigh teen inches, and from one to three inches in diameter. Usually rounded at both ends, there are some which expand at the lower end,2 thus affording a circular, flat crushing surface. The tipper ends of others are ornamented--being sculptured after the similitude of the head of a bird, animal, or snake, and sometimes in imitation of the male organ of generation.
Figures 6 and 8, Plate XVIII., represent the cus tomary forms of maize-crushers or triturating stones. Kelics of this class are very numerous. They are gen erally circular in form, with two flat surfaces, or one flat and the other convex, and can be conveniently grasped and manipulated with the hand. The flat surfaces plainly indicate the use to which they were applied. Sometimes round, water-worn pebbles were employed as mealing-stones--no pains having been
1 u Explorations," etc., vol. ii., p. 370. New York, 1854. a Compare "Ancient Monuments of the Mississippi Valley," p. 220, rig. 118. Washington, 1848.
PESTLES.----MAIZE--CRL'SHERS.----HUT--STONES.
315
taken to modify their natural shapes where they could lie made serviceable. Diorite, quartz-rock, agate, and flint, -were the favorite materials from which these pestles and maize-crushers were manufactured.
In this connection, it seems proper that we should notice a class of relics found in considerable quantities in Middle and Upper Georgia. When I first observed them upon the site of an ancient Indian village near the confluence of Great Kiokee Creek and the Savannah River, I was somewhat at a loss to comprehend their precise use. More than thirty were there seen -within the space of a few acres. They consist of irregular masses of compact sandstone or soap-stone, -weighing from two to ten pounds, in -whose surfaces occur cir cular depressions from an inch to an inch and a half in diameter, and from one-quarter to three-quarters of an inch in depth. Upon the broadest and flattest sides, these depressions, from three to five in number, are located close together. (See Fig. 7, Plate XVIII.) To produce them the harder stones had been pecked, and the softer, gouged. Not only on one side do they ap pear, but frequently on both sides and often in the ends, so that the stone, when set up in the earth on any one of its faces, would always present one or more of these cup-shaped cavities, ready for use.
The Gentleman of Elvas * informs us that in Chiaha, "There was abundance of lard in calabashes, drawn like olive-oil, which the inhabitants said was the fat of bear. There was likewise found much oil of wal nuts, "which, like the lard, was clear and of good taste."
Biedma * confirms this statement, and says, " In this
! " Narratives of the Career of Hernando de Soto," translated by Buckingham Smith, p. 63, Ne- York, 1SS6.
! Ibid., p. 241. New York, I860.
316
ANTIQUITIES Of THE SOUTHERN INDIANS.
province where we "began to find the towns set about with fence, the Indians get a large quantity of oilfrom ivalmcts." -A_t various points reached during the prog ress of the expedition, -walnuts were fonnd stored in the granaries of the natives; and Cabec a de Vaca as serts that these nuts, ground with a small kind of grain, furnished subsistence for two months in the year.' Under the term walnut, the historians of the expedition probably included not only the nut which we designate by that name, but also all the varieties of the hickory-nut with which the country abounded. It is clear that in his forty-fourth chapter the Knight of Elvas confounds the pecan-nut -with the walnut.' " Westward of the Rio Grande," says he, " the walnut differs from that which is found before coming there, being of tenderer shell, and in form like an acorn: while that behind, from the river back to the port of Espiritu Santo, is generally rather hard, the tree and the nut being in their appearance like those of Spain." .Among the Indians of Louisiana so important an article of food was the walnut, that the thirteenth moon was called the 'walnut moon. It was during that month that they cracked their nuts to make bread of them by mixing them with the flour of maize." Bernard Ro mans 4 assures vis that the Florida Indians used Jiiclcorynuts in plenty, making from them a milky liquor of which they were very fond, and which they ate with sweet potatoes." From Bartram's Travels," we extract
1 " Relation of Alvar Nuiicz Cabesa de Vaca," translated by Buckingham Smith, p. 90. Now York, 1811.
3 " Narratives of the Career of Hernando de Soto," translated t>y Buckingham Smith, p. 202. New Tork, 1866.
3 Da Pratz.' " History of Louisiana," vol. ii., p. 195. London, 1*763. 4 " A Concise Natural History of East and West Florida," etc., p. 68. New York, llli. 5 Page 38. London, 1762.
USE OF HICKORY--NUTS AND WALNUTS.
317
the following : " We then passed over large, rich savan nas, or natural meadows, wide-spreading cane swamps, and frequently old Indian settlements, now deserted and overgrown with forests. These are always on or near the banks of rivers, or great swamps, the artificial mounts and terraces elevating them above the surround ing groves. I observed in the ancient cultivated fields : 1. Diospyros ; 2. Gleditsia triacanthos ; 3. Primus cliicasau; 4. Callicarpa; 5. JVtorus rubra; 6. Juglans exaltata; 7. Juglans nigra, which inform us that these trees were cultivated by the ancients on accoimt of their fruit as being wholesome and nourishing food. Though these are natives of the forest,1 yet they thrive better, and are more fruitful in. cultivated plantations, and the fruit is in great estimation with the present generation of Indians, particularly Juglans exaltata, commonly called shell-barked Mccory. The Creeks store up the last in their towns. I have seen above an hundred bushels of these nuts belonging to one family. They pound them to pieces, and then cast them into boiling water, which, after passing through fine strainers, preserves the most oily part of the liquid; this, they call by a name which signifies hiccory milk; it is as sweet and rich as fresh cream, and is an ingredient in most of their cookery, especially homoiiy and corn cakes." Inferring to the use made . of walnuts by the Virginia Indians, Harlot writes: " Besides their eating of them after our ordiiiarie mauer, they breake them with stones, and pound them, in morters with "water to make a milk which they vse to put into some sorts of their spoonmeate ; also among
318
ANTIQUITIES <XF TUB SOUTHERN INDIANS.
their soclde wheat, peaze, beanes, and pompions which inaketh them haue a farre more pleasant taste." '
We have thus, at some length, referred to the use of nuts as an article of food among the Southern In dians, because we hence derive the meaning- and em ployment of these cup-shaped cavities. In our judg ment these relics are simply the stones upon which the Indians cracked their nuts. Their cavities are so located that one, two, three, four, five, and sometimes more nnts could be cracked at a single blow delivered by means of the circular, flat crushing-stone so com mon, and so often found in direct connection with the rude articles now under consideration. The cups are just large enough to hold a hickory-nut or a walnut in proper position so that, "when struck, its pieces would be prevented from being widely scattered. Particu larly do the soap-stones indicate the impressions left by the convex surfaces of the harder nuts. Upon some of them the depressions seem to have been caused simply by repeatedly cracking the nuts upon the same spot so that in time a concavity was produced corre sponding to the half of the spherical or spheroidal nut. Such is the most natural explanation we can otFer with regard to the use of these stones.
In one of the "Western mounds Messrs. Squier and Davis found a block of compact sandstone, weighing between thirty and forty pounds, with several circular depressions resembling those in the work-blocks of coppersmiths in which plates of metal are hammered to give them convexity. These depressions were arti ficial, and possessed various diameters. It was suggested
1 "A Briefe and True Report," etc., p. 18. Franeoforti ad Mcenum anno 1590.
STOCKS TJPOIT "WHICH iNTJTS WERE CKACICED. 319
that in such moulds disks or medals of copper were formed. 1
Colonel Charles "VVhittlesey, in a recent monograph,2 alludes to tlie existence of hundreds of stones, in Cuyahoga Valley and throughout the northern portion of Ohio, with circular, cup-shaped cavities, sometimes on one side and again on l>oth sides, with diameters varying from a point to an inch and a half, and half the diameter in depth, which, from the description given and from the photograph of one of them, we are inclined to regard as very similar to, if not identical with, those which have just engaged our attention. He pronounces them spindle-socJcet stones. Without a personal inspection of these relics it would not l>e proper to express a decided opinion ; and yet, in view of the facts as they appear, we cannot resist the im pression that these too are stones on which nuts were cracked by the primitive peoples who dwelt in the rich valleys of Ohio. It comports not with our present design to criticise the suggestions of Messrs. Squier and Davis, and of Colonel \Vhittlesey, with re gard to-the particular specimens which claimed their examination ; nevertheless, I am free to confess, while standing upon the sites of ancient Indian villages, in Georgia, at present overshadowed by large hickory and walnut trees filled with fruit--calling to rniud the re corded observations of the early travellers concurring in the statement that the red-men of this region in dustriously collected and hoarded these nuts, using them as a favorite article of food in connection with their corn-bread and hominy--conjecturing the method
320
ANTIQUITIES OF THE SOUTHERN INDIANS.
in all likelihood adopted by tliese tribes in crushing them in order that they might conveniently avail them selves of the rich oil and. sweet flavor which, dwelt within, the tough shells--and, upon the very spots where they had long since been abandoned, unearthing these irregularly-shaped stones with their cup-like cavi ties, I felt persuaded that I saw before me physical proofs of the truth of history, and discerned in the lo cality, in their numbers and in the peculiar conforma tion of these rude objects, the purpose they subserved in the olden time.
f
CHAPTER XIV.
BEFOKE the axe of the European was lifted against
the primeval trees, or that system of drainage and de
nudation inaugurated "by which large tracts of densely-
wooded lands have been gradually converted into culti
vated fields and the pleasant sites of cities and villages,
swamps, meadows, and forests, abounded with game of
every description native to this semi-tropical region.
Amid the general silence "which then reigned unbroken,
save by the voices of Nature and the occasional dances,
festivities and war-whoops of the aborigines, there was
little to terrify the wild animals at sport or pasture,
scarcely any thing to affright the birds from their ac
customed homes. The Indian population--limited at
best and confined to chosen seats--was characterized
by remarkable taciturnity. On every hand the air
was vocal with the variant notes of the feathered tribe,
I
and every brake was alive with the forms of animal
life. Buffaloes, bears, deer, cougars, wild-cats, raccoons,
opossums, beavers, rabbits, squirrels, and other quadru
peds, frequented the woods and congregated thickly in
the moss-clad margins "which environed the sluggish
lagoons, undisturbed save by the noiseless yet fatal
32li
ANTIQUITIES OF THE SOUTHEBISr INDIANS.
arrow of the red hunter, uninterrupted in tlieir daily ranges except by occasional villages scattered here and there at long intervals throughout this vast domain.
The buffalo long since ceased, to exist in this region. But few streams give present token of the industry of the beaver. Bears confine themselves to the vinecovered depths of unfrequented swamps. The cry of the cougar is seldom heard in the night-watches. The wolf is no longer a pest, and from whole districts the deer has been expelled. For the untamed denizens of the forest, agriculture and civilization have made no reservations. Expatriation and death have been meted out oven to the hunter-tribes; and they, too, are dwell ers here no longer. In that ancient time, however, there was no lack of food either in the woods or in the waters. The early narratives frequently mention pres ents of deer, bears, and wild-turkeys, at the hands of the Indians, and perpetuate the admiration of the Europeans as they beheld, for the first time, the pathloss forests teeming with game. " The Indians never lack meat," says the Fidalgo of Elvas. ""With arrows they get abundance of deer, turkeys, conies, and other wild animals, being very skilful in killing game."' " They are excellent Hunters," affirms Thomas .Ash, " their Weapons the Bow and Arrow made of a Read pointed with sharp Stones or Fish Bones." 2 Stillhunting was the favorite style, and in plate xxv. of the " Brevis ^Narratio " we have a quaint picture of three Florida Indians who, concealed in the skins of stags, and with drawn bows in their hands, have crept upon
dri Soto," etc., p, 05. Translation London, 1682.
IIUSTISG THE KOE--DEER.
323
and are on the eve of discharging their barbed arrows into a herd of deer drinking at a stream.'
Bossu thus describes the method adopted by the Alibamons in hunting the roe-deer : " Au Indian takes the head of a roe-buck and dries it; he then carries it with him into the woods, where he covers his back with the skin of this animal; he puts his hand into the neck of the dried head, taking care to put little hoops under the skin to keep it firm on the hand; lie then kneels down, and in that attitude, mimicking the voice of these creatures, he shews the head ; the roe-deer are deceived by it and come very near the hunters, who are sure to kill them."
As the woods were well stocked with game, so also was there plenteous supply of fishes in ponds, lakes, rivers, and arms of the sea. Depending for subsistence upon wild animal, bird, and fish, the natives "were com pelled to devote most of their time to hunting and fish ing. Certain seasons were entirely set apart to these pursuits, and with formal ceremonies and solemn invo cations were the general expeditions in o^iest of game inaugurated. With no domesticated animal except the dog, they were not entirely improvident of the future. Public, granaries a there "were, iu which were carefully stored the gathered corn and native fruits. At the appointed moons 4 the men assembled for hunting and fishing, often departing upon long journeys, and return ing laden with well-dried meat and the skins of the slain.
1 " Bi-evls Nan-atio," etc., plate xxv. Franeoforli ad Hfcnum. Pe Bry, anno 1591.
2 "Travels through Louisiana," vol. i., p. 259. London, 1Y71. 3 "Brevis Nurratio," plate xxii. * Generally toward the end of October. Bossu'a " Travels through Louisiana," vol. i., p. 259. London, 1V71.
324
ANTIQUITIES OJ? THE SOTJTHEBN INDIANS.
In that remote period when rivers and bays "were navigated only by light cypress canoes whose paddles scarce caused a quiver among the pliant reeds which fringed tlieir banks, when every pond and swamp was fenced in by robust trees and penetrated by huge roots and fallen trunks affording ample protection to the finny tribe, the waters, one and all, were doubtless far more replete with animal life than they are at the pres ent time. The appetites and the more skilful contriv ances of a superior and a denser population, the de struction of forests, the drainage of natural reservoirs, and the noises of commerce, have tended materially to diminish the supply of fish. So plentiful were the fishes in the ponds and shallow puddles which were encoun tered along the line of De Soto's inarch, that they were readily killed with cudgels. The captive Indians, who, in chains, -were compelled to accompany the expedition, while floundering through these lagoons, so disturbed the mud at their bottoms, that the "fish becoming stupefied, would swim to the surface, when as many were taken as were desired." *
Eibault says, as he ascended a goodly and great river on the Florida coast, he found its waters " boiling and roaring through the multitude of all kinds of fish."
For three .or four months in the year the Indians resorted to the coast and subsisted mainly upon oys ters." Tribes inhabiting the interior, when in the spring the shad were running up the Savannah and other Georgia rivers, would encamp upon the bluffs, and, during the continuance of the season, devote
1 " Narratives of the Career of Hernando de Soto," p. 121, translated by Buckingham Smith. Kew York, 1866.
3 " Relation of Alvar Nuuez Cabe9a de Vaca," translated by Buckingham Smith, p. 78. New York, 1871.
VARIOUS METHODS OF CAPTURING FISH.
325
themselves almost exclusively to the capture of these fishes. The unios and various mussels of the fresh water streams were eagerly collected and opened with a view to securing the pearls ' which they contained, and for the purposes of food. Physical proofs of the habits of the natives in this regard remain to the pres ent day. Some of the islands and headlands along the coast are dotted all over with kitchen-refuse-piles in which the shells of oysters, clams, and conchs, largely predominate. Extended artificial deposits of a similar character, composed of fresh-water shells, are still ex tant in many localities where the flow of the river is so interrupted by rooks or shallow places as to fur nish opportunity for the facile construction of wears, or permit the eager sportsman to spear the fishes as they loitered in the eddies or concealed themselves be neath the shadows of the bowlders rising above the level of the brawling current. la these refuse-piles-- the accumulation of centuries--bones of large fishes abound, and net-sinkers are not infrequent.
It is interesting to note the various methods em ployed by the aborigines for the capture of fish.
We have the authority of the Knight of Elvas for the statement that jisJi-preserves existed among the Southern Indians.
When De Soto entered Pacaha, he quartered him self in the town where the cacique was accustomed to reside. It was enclosed and very large. In its towers and palisade were many loop-holes. Much dry maize had been there accumulated, and the new in great quantity was growing in the adjacent fields. !Near the enclosure was " a great lake, and the water entered a
1 Garcilasso de la Vegii, " Conquete de la Floride," trad, par Richelet. Leide, 1731, tomft i., livre 2, chap, i., p. 296, et tteq.
320
ANTIQUITIES OF THE SOUTHER!?1 INDIANS.
ditch that well nigh went round tlie town. 1 From the River Grande to the lake was a canal through which the fish came into it, and where the Chief kept them for his eating and pastime. W'itJb nets tliat 10ere found in the place as many were taken as need, required; and, however much might "be the casting, there was never any lack of them. In the many other lakes about were also many fish, though the flesh "was soft, and none of it so good as that which came from the river." 2
We have here, as has already been suggested, a probable explanation of the principal purpose the reservoirs and the ditch surrounding that remarkable group of mounds near the Etowah River, on Colo nel Tumlin's plantation, were designed to subserve. Through the mouth of that canal fishes could readily enter from the river. Once in, that mouth, could have been closed by means of a wicker-work of cane or split wood so as to prevent their escape. Thus introduced into these artificial lakes, by a very simple contrivance they could be there detained, fed, multiplied, and kept ready for daily use. By means of nets they could be fished out as occasion required. If it be true, as wehave surmised, that the large tumulus was a temple of the sun, it may be that this canal and these lakes were at great labor constructed as fish-preserves for the par ticular benefit of the priests who ministered and the devotees who worshipped there. Similar arrangements for pisciculture are still to be seen in other localities within the present geographical limits of Georgia.
3fish.iT.ig with hook and line seems to have obtained. to a very limited extent, if we may judge from the re-
a plain, well fenced about, and sur-
VARIOUS METHODS OP CAPTURING FISH.
32^7
markable absence of any thing like bone, flint, and shell hooks in the mounds and refuse-piles. Very few hooks have been found, so far as our information ex tends, and they were made of bone.
Fishes were often captured by means of a bright fire, 1 kindled in the canoe which, was paddled by night over their feeding-grounds. Frightened, blinded, and at the same time attracted by the light, they leaped toward it, and in doing so frequently fell into the boat. This mode was particularly successful on the coast, and those who are familiar with the customs of that region will bear witness that to this day many mullets are caught in this manner by negroes carrying torches in their cypress canoes.
Of the Indians inhabiting to the. south of Florida, it is said: a "Besides their enjoyment of the "water, the natives take abundance of mullet from it, bream and other fish that "breed there, as -well as kinds more numerous that ascend from the sea. They come over the bar, by the mouth, in the season proper to them for casting their spawn, remaining to sport in fresh water until about summer, when the rivei- goes down. This is the principal fishing season. "Then the people of the towns, bringing great bundles of bushes, gather about the holes and pools and beat the water, "when the fishes in the depths becoming intoxicated from the sap, ascend to the surface and are taken. Persons receive no harm from the poison in eating them." This method of intoxicating fishes by pound-
1 Losldel says: 1! In Carolina the Indians frequently use fire in fishing. A certain kind of fish will even leap into the boats which have fire in them."-- (" History of the Mission of the United Brethren," etc., p. 95. London, 1794.)
2 " Halation of Alvar Nuiiez Cabepa do Vaea," translated by Buckingham Smith, p. 181. New York, 1871.
328
ANTIQUITIES OE THE SOUTHERN IHTXEAHS.
ed horse-chestnnts and various roots was extensively adopted by the Southern tribes.'
A favorite and manly mode of taking fish was with the bow and arrow, and with the dart or spear. This savored of sport, and afforded ample opportunity for the display of skill. Father Hennepiu pays the follow ing compliment to the dexterity of the Southern In dians dwelling " upon the Hiver Meschasipi." They " are very snbtil and have such lively and piercing Eyes that tho' the Fishes glide very swiftly in. the Waters, yet they fail not to kill them with their Darts, which they vigorously thrust a little before into the Water when they shoot out of their Bow. Moreover, they have long Poles with sharp Points which they dart from them with great Accuracy, because of their being so sharp sighted; they also kill great Sturgeons and Trouts, which are seven or eight foot under AVater." 2
BartramJ1 gives an account of the capture of a salmon-trout weighing about fifteen pounds, by a young Indian. " The Indian," says he, " struck this fish with a reed harpoon, pointed very sharp, barbed and hardened by the fire. The fish lay close under the steep bank, which the Indian discovered and struck with his reed; instantly the fish darted off with it, while the Indian pursued, without extracting the har poon, and with repeated thrusts drowned it and then dragged it to shore."
Lawson" declares that the hunters of the interior were very expert in striking sturgeon and rock-fish or bass when they came up the rivers to spa,wn; and to
1 See Adair's " History of the American Indians," p. 403. London, 17Y5. 3 " A Continuation of the New Discovery," etc., p. 102. London, 1698. 3 " Travels through North and South Carolina, Georgia," etc., p. 44. London, 1192. 4 " History of Carolina," p. 339. Raleigh reprint, 1860.
VAK1OUS METHODS OF CAPTURING FISH.
329
Dr. Brickell '--the plagiarist--we are indebted for tlie ensuing mention of this particular manner of fishing as practised by the Carolina Indians : " They have Fisli-giys that are made of the Reeds or Hollow Canes / these they cut and make very sharp, "with two Beards, and taper at the Point like a Harpoon; being thus provided, they either wade into the AVater, or go into their Canoes and paddle about the Edges of the Hivers or Creeks, striking all the Fish they meet with in the depth of five or six Feet Water, or as far as they can see them ; this they commonly do in dark, calm Nights, and whilst one attends with a Light made of the Pitcli-pine, the other -with his Fish-gig strikes and kills the Fish. It is diverting to see them fish after this manner, which they sometimes do in the Day; how dexterous they are in striking, is admirable, and the great Quantities they kill by this Method."
Lawson states that the " Indian boys go in the night, and one holding a liglitwood torch, the other has a bow and arrows, and the fire directing him to see the fish, he shoots them with the arrows ; and thus they kill a great many of the smaller fry, and some times, pretty large ones." *
In plate xiii. of the " Admiranda Narratio" six Virginia Indians are represented wading ill the water and busily engaged in spearing fish. Three are dis covered in plate iv., in successful pursuit of a school of fishes, while others in canoes are similarly occupied. Plate xxxvi. of the " Brevis Narratio " assures us that the Florida Indians were addicted to the same sport.
In 1805 Barker observed the Chickasaws ia Duck !River, pursuing, in their canoes, the large fishes which
1 "Xatural History of North Carolina," etc., p. 365. Dublin, 1737. 3 "History of Caroliua," etc., p. 341. Raleigh reprint, 1860.
330
ANTIQUITIES OI? THIS SOUTHERN IHDIAHS.
swarmed in that stream, and taking great numbers of them with spears made of tlie long' canes which grew ill the river-bottoms. These spears, says the narrator, " were sixteen, or eighteen feet in length, sharpened with a knife into a lancet shape at one end, and thrown with great 'dexterity twenty or thirty feet; seldom failing to pierco a fish throiigh at every throw. This was doubtless an invention of great antiquity, and practised by their fathers ages before the use of iron was known amongst them."--(" American Pioneer," vol. 1, p. 143. Cincinnati, 1844.)
It was upon their wears, traps, set-nets, and me chanical contrivances of these sorts, however, that the natives largely depended for a constant and liberal supply of iish. Their use, in some form or other, was general. Captain Eibault informs us that the Indians of May ]River put as presents into his boats " sundry fishes which with mervelons speed they run to take in their packs mad e in the water -with great reeds, so well and cunningly set together after the fashion of a Labaryiithe, or Maze, with so many turns and crooks as it is impossible to do it without much consideration and industry." * The Carolina Indians are said to have taken the sturgeon in snares such as are used in E nrope for the capture of pike. " The herrings," according to Surveyor-General Lawson, " in March and April run a great way up the rivers and fresh streams to spawn, where the savages make great wares with hedges that hinder their passage, only in the middle where an arti ficial pond is made to take them in so that they can not return. This method is in use all over the fresh
FISH--TRAPS OR WEARS,
331
streams to catch trout and the otlier species of" fish which those parts afford." *
Dr. Brickell a is rather more definite in his descrip tion, and advises us that these wears were constructed of " long poles or hollow canes." 3
In plate xiii. of the " Admiranda Narratio," we find a distinct representation of one of these fish-traps, with extended wings; one of which reaches the shore, and the other far out into the -water. It is made of canes or small poles firmly stuck in the mud, so as to pre serve an upright position. Placed close to each other, and rising a few feet above the water-level, they are securely fastened together by parallel ropes or withes, thus forming a sort of hedge or rustic fence through, which the fishes are unable to force a passage. In the middle is an opening leading into a circular enclosure. This, by a circuitous opening, communicates with a second pen, and this in like manner with a third, and that, in turn, in a similar way with a fourtli--each somewhat smaller than the former.* Two Indians are seen in a canoe at the opening of the wear. The one in the bow \vith a scoop-net is dipping up the fish in the first pen, while numerous other fishes are figured making their way into the other enclosures, whence, for them, there can be little or no hope of escape. The explanatory text is as follows.: "Egregiam etia habent piscancli in fiuminibus rationem : cum enim ferro & chary be careant, arundinibus aut oblongis virgis piscis cniusdam cancro mariiio similis cauclam concauam pro
1 History of Carolina," etc., p. 339. Raleigh reprint, 18GO. a "Natural History of North Carolina," etc., p. 366. Dublin, 173*7. 3 " Cabega de Vaca mentions wears made of cane." Translation of Backing- , ham Smith, p. 75. New York, 1871. 4 Here we have an explanation of what Captain Ribault calls " a Labarynthe or Haze with so many turns and crooks,"
332
ANTIQUITIES OF THE SOUTHERN INDIANS.
cuspide imponunt, quibus nocta vel intertliu pisces figunt, &, in suas cymbas congerunt: sed aliorum piseium spinis <fe spiculis uti norunt. Baculis etiam seu virgultis in aquam defixis tegetes conficiuut, quas intertexentes in angustum semper coiitrahunt, ut ex figura apparet, nunquam apud. nos conspecta est tana subtilis pisces oapiendi ratio, quorum varia genera istic in fluminibus reperiuntur, nostris dis similia & boni admoduni succi."
By some of tne illustrations accompanying the " Brevis Narratio," ' we are persuaded that similar wears were constructed "by the Florida Indians.
Lostiel 3 describes a particular mode of fishing wliicli was protaWy adopted ill some of the Georgia rivers : " W^lien the shad fish (cltrpea alosa) come up the rivers, the Indians run a clam of stones across the stream, 'where its depth will admit of it, not in a straight line, but in two parts verging towards each other in an angle. An opening is left in the middle for the "water to run off. A.t this opening they place a large box, the bottom of which is full of holes. They then make a rope of the twigs of the wild vine, reach ing across the stream, upon which "boughs of about six feet in length are fastened at the distance of about two fathoms from each other. A party is detached about a mile above the dam w^h this rope and its append ages, who begin to move gently down the current, some guiding one, some the opposite end, -whilst others keep the branches from sinking by supporting the rope in the middle with wooden forks. Thus they proceed, frightening the fishes into the opening left in the mid dle of the dam, where a number of Indians are placed
1 Plato iii. 3 " History of the Mission of the United Brethren," etc., p. 95. London, 1704.
ADAIK'S ACCOTTNT OF INDIAN FISHIXG. '333
on each side, who, standing upon the two legs of the angles, drive the fishes with poles, and an hideous noise, through the opening into the above-mentioned box: or chest. Here they lie, the -water running off through the holes in the bottom, and other Indians stationed on each side of the chest, take them out, kill them and fill their canoes. By this contrivance they sometimes catch above a thousand shad and other fish in half a day."
]Mr. Axlair's summary ' of the various methods ad opted by the Southern Indians, and particularly the Georgia tribes, in their practice of the piscatorial art, is so interesting, minute, and appropriate, that we make no apology for repeating it in exten&o:
"Their method of fishing may be placed among their diversions, but this is of the profitable kind. "When they see large fish near the surface of the water, they fire directly upon them, sometimes only with pow der, which noise and surprize, however, so stupifies them that they instantly turn up their bellies and float atop, when' the fisherman secures them. If they shoot at fish not deep in the water, either with an arrow or bullet, they aim at the lower part of the belly, if they are near; and lower, in like manner, according to the distance, which seldom fails of killing. In a dry sum mer season, they gather horse-chesnuts and different sorts of roots, which, having pounded pretty fine, and steeped a while in a trough, they scatter this mixture over the surface of a middle-sized pond, and stir it about with poles, till the water is sufficiently impreg nated- with the intoxicating bittern. The fish are soon inebriated and make to the surface of the "water, with their bellies uppermost. The fishers gather them
1 " History of the American Indians," pp. 402-405. London, 1765.
334
ANTIQUITIES OF THE SOUTHERN INDIANS.
in baskets, and barbieue the largest, covering them carefully over at niglit to preserve them from the sup posed putrifying influence of the moon. It seems that fish catched in this manner are not poisoned, but only stupified ; for they prove very wholesome food to us, who frequently use them. By experiments, when, they are speedily moved into good water, they revive in a few minutes.
" The Indians have the art of catching fish in long crails, made with canes and hiccory splinters, tapering to a point. They lav these at a fall of water, where stones are placed in two sloping lines from each bank, till they meet together in the middle of the rapid stream, where the entangled fish are soon drowned. Above such a place I have known them to fasten a wreath of long grape-vines together to reach across the river, with stones fastened at proper distances to rake . the bottom : they will swirn a mile with it whooping and plunging all the way, driving the fish before them into their large cane pots.' "With this draught, which is a very heavy one, they make a town feast, or feast of love, of -which every one partakes in the most social manner, and afterward they dance together, singing Halelti-yali, and the rest of their usual praises to the divine essence, for his bountiful gifts to the beloved people. Those Indians who are unacquainted with the use of barbed irons, are very expert in striking large fish out of their canoes, with long sharp-pointed green canes, which are well bearded, and hardened in the fire. In Savannah River I have often accompanied them in killing sturgeons with those green swamp harpoons, and which they did with much pleasure and ease; for when we discovered the fish, we soon thrast
1 See also " Jlemoira of Lieutenant Ti'mberlake," p. 43. London, 1765.
ADAIB'S ACCOUKT OF ISDIAU FISHING. 33o
into tlieir todies one of the harpoons. As the fish would immediately strike deep, and rush away to the bottom very rapidly, tlieir strength was soon expended by tlieir violent struggles against the buoyant force of the green darts: as soon as tlie top end of them ap peared again, on the surface of the water, we made up to them, renewed the attack, and in like manner con tinued it till we secured our game.
" They have a surprising method of fishing under the edges of rocks that stand over deep places of the river. There, they pull off their red breeches, or their long slip of Stroud cloth, and wrapping it round tlieir arm, so as to reach to the lower part of the palm of their right hand, they dive under the rock where the large cat-fish lie to shelter themselves from the scorching beams of the sun, and to -watch for prey : as soon as those fierce aquatic animals see that tempting bait, they immediately seize it with the greatest violence in. order to swallow it. Then is the time for the diver to improve the favorable opportunity : lie accordingly opens his hand, seizes the voracious fish by his tender parts, hath a sharp struggle with it against the crevices of the rock, and at last brings it safe ashore. Except the Choktah, all our Indians, both male and female, above the state of infancy, are in the -watery element nearly equal to amphibious animals, by practice : and from the experiments necessity has forced them to, it seems as if few -were endued -with such strong natural abilities--very few can equal them in their wild situa tion of life.
" There is a favorite method among them of fishing with hand-nets. The nets are about three feet deep, and of the same diameter at the opening, made of hemp, and knotted after the usual manner of our nets.
336
AMTIQTJITIES OF THE SOTJTHEBJT IHTIIA^S.
On each side of the mouth tliey tie very securely a strong elastic green cane, to wMch the ends are fast ened. Prepared with these, the warriors abreast jump in at the end of a long pond, swimming tinder water, with their net stretched open with both hands, and the canes in a horizontal position. In this manner they will continue either till their breath is expended by the waiit of respiration, or till the net is so ponder ous as to force them to exonerate it ashore, or in a basket fixt in a proper place for that purpose--by removing one hand the canes instantly spring together. I have been engaged half a day at a time with the old friendly Chikkasah, and half drowned in the diversion --when any of us was so unfortunate as to catch watersnakes in our sweep, and emptied them ashore, -we had the ranting voice of our friendly posse comitatus, "whoop ing against us till another party was so unlucky as to meet with the like misfortune. During this exercise the women are fishing ashore with coarse baskets, to catch the fish that escape our nets. A-t the end of our friendly diversion, we cheerfully return home, and in an innocent and friendly manner eat together, studious ly diverting each other oil the incidents of the day, and make a cheerful night."
It appears that the Southern Indians were fond of crawfish as an article of food. Selecting a stream fre quented by such fishes, they angled for them in the following manner : Slips of half-roasted or barbecued venison were strung, about six inches apart, upon reeds sharpened at one end. Thus baited, a great many of these reeds were stuck in the bed of the brook. Re maining near, the Indians watched these baited reeds, pulling them up at intervals, shaking into baskets the crawfish adhering to the bits of meat, and then re-
UET--SINKERS AND FISHING--PLUMMETS.
33*7
placing them in tlie water. " By this method," says Lawson, " they will ill a little time catcli several bushels." Blackmoor's teeth were taken in great quantities "by means of oysters tied to strings. The coast Indians, carrying them into the interior, traded them away to remote tribes by whom they were held in much esteem. 1
In addition to the modes already enumerated, it may be safely asserted that nets were also used .by the natives for the capture offish. Of their peculiar shape and construction we have no specific account. Remem bering, however, the ingenuity displayed by these peo ples in the fabrication of garments from the fibres of trees, mats from rushes, and ornamental coverings from feathers, it would be singular if in the silk-grass, the inner bark of the mulberry, and. other natural sub stances of this region, they had not found convenient materials for the manufacture of substantial nets and lines. Their former existence is indicated by the pres ence of SINKERS and FISHING-PLUMMETS.
Of the sinkers, there are two varieties--perforated, and notched or grooved. Regarding them as a whole, we may state that they were usually made of soapstone, sometimes of slate, rarely of flint or hard stone, and occasionally of clay. All of the perforated sort that I have seen, with one exception, were formed either of soap-stone or of clay. Consisting generally of flat or rounded pieces of soapstone irregular in shape, they vary in weight from scarcely more than an ounce to a pound and upward. The perforations are from a quarter of an inch to an inch in diameter, and are indifferently located either in the centre or near the'cdge of the stone. Of this variety, Figs. 1, 2, 3,
1 &e "*La\vson's History or C.'irolina," p. 340. Raleigh reprint, 18GO. Brickell's " Natural History of North Carolina," p. 3C7. Dublin, 1T3Y.
338
ANTIQUITIES OF THE SOUTHERN' INDIANS.
4, and 5, Plate XIX., may be taken as typical speci mens. In tliis connection we would refer to an. object (Fig. 6, Plate XIX.) of soap-stone, well worked in every part, which was found in a relic-bed on Price's Island, in the Savannah River, opposite Columbia County, and associated -with several perforated net-sinkers of similar material. It is eight inches and a half in length, six inches and a half broad at the widest part, and about three-quarters of an inch in thickness. The perforation is three-quarters of an inch in diameter. It is suggested that this article should be classed with net-sinkers. Little labor was bestowed upon the manu facture of the notched sinkers (Figs. 7, 8, and 11, Plate XIX.), the only object being to rudely break or chip the material into convenient size, and then notch it at the opposite sides or ends so that it could be securely attached by means of a vine, a strap of deer-skin, or a thong of some kind. Such plummets are, as a rule, bulky, and were probably used to weigh down the long grape-vine ropes 1 -with' which the Indians were wont to drag the rivers in driving the fishes before them into their large cane traps. The noise of these stones rolling along the bottom would have materially assisted in frightening the fishes fi-oin their hidingplaces and in compelling them to swim toward the de sired point.
Other fishing-plummets 2 (Fig. 10, Plate XIX.) have a single groove around the middle, while others still (Fig. 9, Plate XIX.) have two or more grooves inter-
1 See Adair's " History of the American Indians," p. 403. London, 1775. 2 Prof. Rau has in his collection net-sinkers, notched and grooved, found near Sfuncy, on the banks of the Susquehanna. River. I have seen similar ones from the shores of Rhode Island. These types are also represented in the islets and reefs of the. west coast of Sweden (Nilssou's " Stone -Age," p. 26, plate ii., Figs. 32, 34, 35, London, 1868), and in other localities in Europe.
NET--SINKERS.----NETS.
339
secting each other at right angles. These grooves are carelessly cut or pecked, and are intended to facilitate the attachment. Fig. 12, Plate XIX., illustrates a more carefully-wrought kind of plummet, which may have been employed to weight the hand-line in fishing ^vith a hook.
These sinkers abound along the banks of the Savan nah River above Augusta, and. are found upon the bluffs of other streams where the Indians habitually congregated for the purpose of fishing. Near the con fluence of Great Kiokee Creek aad the Savannah River an extended kitchen-refuse-pile was cut in two and laid bare, some years since, by a heavy freshet. Hundreds of these perforated and notched sinkers were there unearthed, showing the great quantities manufactured and used by the natives at this point.
In his account of the fish-preserve near the village of the Cacique of Pacaha, the Gentleman of Elvas inti mates that cast-nets' were there made and used by the natives. Cabeca de Vaca, on more than one occasion, alludes to the existence of nets, and it may be that the smaller kinds both of the perforated, and grooved or notched plummets, served as net-sinkers. It is not improbable that the Southern Indians manufactured and fished with set or gill nets, which would have proved very effective in the capture of shad. In that event these large plummets would have answered, well as weights to keep the nets in proper position. In plate xiii. of the " Admiranda Narratio " two forms of nets are figured; one, the ordinary dip or scoop net, and the other, conical-shaped, its apex terminating in a long handle. The latter was made of cane or split
1 "Narratives of the Career of Hernando de Soto," translated by BLickingham Smith, p. 112. New York, I860.
340
ANTIQUITIES OF TIIE SOUTHERN INDIANS.
wood--the longitudinal ribs, after leaving the handle, expanding at the bottom, where they were kept in place by means of circular and parallel cords or hoops of cane, thereby forming a stiff enclosure, open beneath, which could be thrust over a fish or crab in shallow water. Thus detained, tlie animal could not escape, and was subjeet to immediate manucaption. Nowhere, so far as our personal information goes, have either cast-nets or trolling-nets been particularly mentioned or described by the early narrators. In the present state of the inquiry, it does not become us, however, to say that such nets were not in use among the Southern Indians. The probability is that they did have some such contrivances. AVe would suggest, nevertheless, that the principal office of the plummets and sinkers we have been examining was either to assist in steady
ing and anchoring the traps, or to act as weights for set-nets ; or, what is most likely, to carry to the b'ottom the long grape-vine ropes "with which the natives dragged the streams when they wished to rout the fishes from their lurking-places and drive them into their cane labyrinths or wears.
Many of these sinkers consist simply of water-worn pebbles or irregular fragments of rock rudely notched around the centre, and sometimes longitudinally also. Little labor was expended save in the selection of stones of proper sizes, and in pecking such grooves as would permit secure attachment to the upright poles of the wears, the ends of fishing-lines, and to the short grape-vines depending from the stout mother-vine, with which the aborigines were wont to drag the rivers. Some of the heavier and rougher sort may properly be denominated anchors for wears and stationary nets, or fish-traps.
CHAPTER XV.
Discoidal Stones.--Chungke Game.
IN Iris most interesting and valuable historical sketch of Germany, Tacitus ' mentions the fact that the ancient Grermans were so passionately addicted to a game of chance that, when all their property had been gambled away, the desperate players would hazard upon a final throw even their personal liberty.
AVith almost equal desperation, if we may credit Adair, did the Cherokees pursue their national game of CIITING-KT:. After describing their ball-playing, he states : " The warriors have another favorite game called Ghunylce, which, with propriety of language, maybe called 'Running hard labour.' Tliey have near their state-house a sc.pj.are piece of ground well cleaned, and fine sand is carefully strewed over it, when requi site, to promote a swifter motion to what they throw along the surface. Only one or two on a side play at this ancient game. They have a stone about two fingers broad at the edge, and two spans round ; each party has a pole of about eight feet long, smooth, and tapering at each end, the points flat. They set off
1 "C. Cornolii Taciti Opera onrnia, ad fidem editionis OrcllianEe," torn, i:., p. 243. Oxonii, 1851.
342
AlSTTIQTJITIliS OF THE SOUTHERN INDIAN'S.
abreast of eacli otlier at six yards from tlie eiicl of tlie play-ground ; then one of them hurls the stone on its edge, in as direct a line as lie can, a considerable dis tance toward, the middle of the other end of the square : when they have ran a few yards, each darts his pole anointed with bear's oil, with a proper force, as near as he can guess in proportion to the motion of the stone, that the end may lie close to the stone: when, this is the case, the person counts two .of the game, and, in proportion to the nearness of the poles to the mark, one is counted, unless by measuring-, both are found to be at an equal distance from the stone. In this manner the players will keep running most part of the day, at half speed, under the violent heat of the sun, staking their silver ornament?, their nose-, finger-, and. ear-rings ; their breast-, arm-, and wrist-plates, and even all their wearing-apparel, except that which barely covers their middle. All the American Indians are much addicted to this game, which to us, appears to be a task of stu pid drudgery : it seems, however, to be of early origin when their forefathers used diversions as simple as their manners. The hurling'-stones they use at present were, time immemorial, rubbed smooth on the rocks, and with prodigious labour; they are kept with the strictest religions care from one generation, to another, and are exempted from, being buried with the dead. They belong to the town where they are used, and are care fully preserved." * Physical traces exist to this day, in various portions of Georgia, denoting the carefullyprepared spaces or areas dedicated in the olden time to the uses of this game. These are parallelogrammic in shape, slightly elevated, and are from sixty to ninety feet in length, and about half as wide. When Adair
1 " History of the American Indians," etc., p. 401, et seq. London, 1V75.
DISCOIDAIj STONES.
343
says that these " hurling-stones " were kept with tlie
utmost religious care, fi-om one generation to another,
and were exempt from inhumation with the dead, he
states a fact which was the result of his extended per
sonal observation. His assertion, however, is not en
tirely correct. One of the finest discoidal stones the
writer has ever seen, was taken from a mound in Cass
County, near the Etowah River, about thirty feet below
the upper surface of the tumulus. A similar relic lay
touching it. They had both been placed on edge, and
at right angles to each other. Above them was a
layer of human bones in a decomposed state. From a
small sepulchral mound on Pope's plantation, in the
Oostenaula Valley, we obtained a discoidal stone of
ferruginous quartz, almost the counterpart of those
just alluded to. A little more than a year ago, a
freshet in the Oconee River carried away a portion of
a mound which stood upon its bank, not far from
Athens, and in doing so washed out a discoidal stone
of quartz, hollowed out on both sides to the depth of
an inch, five inches in diameter, carefully polished and
perfect in every particular. These may be exceptional
cases, but they are worthy of note. In view of the
special esteem in which such articles must have been
held, remembering the protracted labor involved in
their manufacture, and mindful of the universal fond
ness cherished by the natives for the same in which
they were thrown, we can readily believe that these
discoidal stones wore carefully preserved, and, because
of their great value, excused from sepulture with the
general dead. If it be true, as some have asserted,
that they were regarded as the common property of a
t
town or community, we have in this circumstance ad ditional reason for supposing that they should have
344
ANTIQUITIES OF THE SOUTHERN INDIANS.
escaped interment, except, perhaps, with, some distin guished person or noted player of the game.
While enumerating the principal sports of the In dians of Georgia and Florida, Captain Bernard Ro mans' makes the following specific mention of the unique and absorbing game whose peculiarities are engaging our present attention : " Their favorite game of cliunke is a plain proof of the evil consequences of a violent passion for gaming upon all kinds, classes and orders of men; at this they play from morning till night with an nnwearied application, and they bet high: here you may see a savage come and bring; all his skins, stake them and lose them ; next his pipe, his beads, trinkets, and ornaments ; at last, Iris blanket and other garment, and even all their arms; and, after all it is not uncommon for tliem to go liome, borrow a gun and shoot themselves. , . . The manner of playing this game is thus : They make an alley of abont two hun dred feet in length, where a very smooth caly ground is laid, which when dry is very hard; they play two together, having each a streight pole o about fifteen feet long; one holds a stone, which is in shape of a truck, which he throws before him over this alley, and the instant of its departure they set off and run ; in running they cast their poles after the stone; he that did not throw it endeavors to hit it, the other strives to strike the pole of his antagonist in its night so as to prevent its hitting the stone; if the first should strike the stone, he counts one for it, and if the other by the dexterity of his cast should prevent the polo of his opponent hitting the stone, he counts one, but shoidd both miss their aim, the throw is renewed:
THE CHUNGKE GAME.
345
and ill case a score is won, the winner casts the stone and eleven is up ; they hurl this stone and pole with wonderful dexterity and violence, and fatigue them selves innch at it."
In' describing the chunk-yards in vogue among the Creeks, Bartram expresses the opinion that they were of very ancient date, and not. the work of the modern Indians. It has been supposed, and apparently with very good reason, that these areas were chiefly devoted to the practice of this favorite game; and that instead of calling them cliiink-yards, we ought properly to denominate them chunglce-yardg*
According to Dn Pratz," the method adopted, by the Louisiana Indians in playing this game differed somewhat from that prescribed among the Indians of Georgia and Florida. " The warriors practice a diver sion which is called the game of the pole, at which only two play together at a time. Each has a pole about eight feet long, resembling a Roman f, and the game consists in rolling a flat round stone, about three inches diameter, and an inch thick, with the edge somewhat sloping, and throwing the pole at the same time in such a manner that when the stone rests the pole may touch it or be near it. Both antagonists throw their poles at the same time, and he whose pole is nearest the stone counts one, and has the right of rolling the stone. The men fatigue themselves much at this game as they run after their poles at every throw ; and, some of them are so bewitched by it that they game away one piece of furniture after another. These gamesters, however?
3-16
ANTIQUITIES OF THE SOUTHERN INDIANS.
are very rare and are greatly discountenanced by the rest of the people."
It would appear, from Lieutenant Timberlake's observations, that this game was also called, among some of the Cherokee tribes, nettecaw^v, of which he gives us the following description : * " Each player having a pole about ten feet long, with several marks or divisions, one of them bowls a round stone, "with one flat side, and the other convex, on which the players all dart their poles after it, and the nearest counts according to the vicinity of the bowl to the marks on his pole."
The Carolina Indians, as we are informed by Sur veyor-General John iLawson,2 were much addicted to a sport they called Chenoo, " which is can-led on with a staff and a bowl made of stone which they trundle upon a smooth place like a bowling green, made for that purpose." The presence of these discoidal stones in Tennessee, Alabama, Mississippi, Louisiana, Ken tucky, Virginia, and elsewhere, assures us that this game of chunglce was generally practised by all the Southern tribes. In his " Natural and Aboriginal His tory of Tennessee," Mr. Haywood describes several stones of this sort, and also declares that they have been found in mounds. Upon page 190 we read : " In the possession of General Cocke, of Grainger County, and in the town of Rutledge, is a circular stone found in the "woods there, of three inches in diameter, resem bling in colour dark yellow barber's soap. In the centre, on each side, is a small circular excavation about one inch in diameter or a little more, scooped out as far as to its circumference, extending not quite
DISCOIDAL STORES.
347
lialf way from the centre to tlie circumference of the stone itself. On both sides there is a declivity from the centre to the edge, making the extremity not more than half as thick as the stone is at the centre. It is very smoothly cnt. . . .
" In the museum of a lady at Nashville is one of a similar shape; it is made of stone, very white like snow, transparent and glittering, very hard and heavy. It is about three inches ill diameter, or perhaps a little less; the excavation in the centre on ea,ch side seems adapted, to the thumb and finger, and at the extremity it is wider in proportion than the one before described. And, lately was taken from a nionnd in Maury Coun ty, a stone perfectly globular, very hard, and heavy, of a variegated exterior and exceedingly well polished. It probably belonged to some employment that the other circular stones did." Again, at page 196, our historian continues: " About ten miles from Sparta, in White County, a conical mound was lately opened, and in the centre of it was found a skeleton eia'ht feet in length. AVith it was found a stone of the flint kind, very hard, with two flat sides, having in the centre cir cular hollows exactly accommodated to the balls of the thumb and forefinger. This stone was an inch and a half in diameter--the form exactly circular. It was about one-third of an inch thick and made smooth and flat for rolling, like a grindstone, to the form of which, indeed, the whole stone was assimilated. When placed upon the floor it would roll for a considerable time without falling. The whole surface was smooth and well polished. . . . No doubt it was buried with the deceased, because for some reason he had set a great value on it in his lifetime, and had excelled in some accomplishment to which it related. The colour
348
ANTIQUITIES OF TH_E SOUTHEEN INDIANS.
of the stone "was a dingy -white, inclining to a darkish yellow." '
"Without multiplying these historical proofs of the presence and use of these discoidal stones among the Southern Indians, even within historic times, we pro ceed to consider briefly the peculiar forms of such as have been found within the present geographical limits of Georgia. It may be stated generally that they are all circular in shape, with diameters varying from one to six inches. In thickness they differ from a quarter of an inch to. two inches and a quarter. Many are flat on the sides, which, as they approach the circumference, become slightly convex. Perpendicular at the edge, they are capable of standing on edge and of maintain ing this upright position, with great tenacity, when rolled along the ground. Others are lenticular in shape, with oblique margins. For the manufacture of specimens of this solid type---which -we presume -was the common form--a hard, black, close-grained stone, capable of receiving a fine polish, formed the favorite material, especially along the coast. So nearly in out line do these frequently resemble the old-fashioned iron weights in use in country stores, that these relics are often spoken of, among the unlearned, as Indian weights. (See Figs. 1, 2, 4, 5, 7, 11, 12,13, Plate XX.) It is probable that the smaller varieties were made for children, who, at an early age, -were taught to imitate this favorite amusement of their elders. This impres sion is strengthened by the fact that numerous disks of pottery with the ornamentation, of the vessel still upon them, are found upon, the sites of old villages and at localities along the river-banks where the natives congregated from year to year to fish. It is
1 " Natural and Aboriginal History of Tennessee," pp. 190, 196. Nashville, 1823.
DISCOIDAL STORES.
349
difficult to conjecture what uses tliese smaller-cliscoidal stories and clay disks were designed to subserve except the training of tlie little ones in the arts and rules of this ancient and universally-esteemed game. Beauti ful varieties of pudding-stone (Fig. 12, Plate XX.), greenstone (Fig. 13, Plate XX.), talcose slate, soapstone, flint, and even agate (Fig. 7, Plate XX.), were employed in tlie manufacture of cliscoidal stones of the solid type. The regularity of outline and the degree of polish are remarkable.
The first modification of this customary shape is seen in those discoidal stones whose sides are slightly concave or convex. In some instances one side appears convex, and the other concave. (Figs. 9 and 10, Plate XX.) ^rVe turn now to the more elaborate forms of these discoidal stones, three of which are represented in the accompanying- plate. They are all made of ferru ginous quartz, and are well polished. The first speci men (Fig. 6, Plate XX.) is evenly hollowed out on both sides to the depth of an inch and a quarter in the centre. The cavities are circular, and four inches in diameter. From the edge of each cavity toward the outer circumference, the stone is bevelled, so that the edge of the disk is just an inch in width. This discoidal stone is five inches and three-quarters in diam eter, and two inches and a half in thickness at the point where the cavities begin.
The second (Fig. 3, Plate XX.), which was taken from a rnound in Bullock County, is somewhat larger than the first, being exactly six inches in diameter, and a little more than two inches and a quarter thick. It has four cavities, two on each side, precisely similar, and one within the other. The diameters of the larger cavities are each four inches and a half; of the smaller,
350
ANTIQUITIES OF THE SOUTHERN INDIANS.
two inches and. a quarter. Tlie depth of the outer cavities is five-eighths of an. incli; of the inner, threeeighths of an inch. As in the first specimen, the sides are here also "bevelled toward the edge, -which is rather more than three quarters of an inch in width. Between the outer circle of the concavity and the point where the bevelling commences, occurs a rim a quarter of an inch in thickness. The third specimen (Fig. 8, Plate XX.), which may justly be regarded as a wonderful illustration of the skill and protracted labor of the primitive artist working without rule or compass and unaided by a single metallic tool, was found, as we have already stated, at the bottom of a large sepulchral mound, thirty feet high, in Cass County. It is abso lutely symmetrical in all its parts, being five inches and three-quarters iu diameter, and one inch and seveneighths in thickness. The cavities--which are precise ly similar on both sides--are three inches and a half in diameter, and three-quarters of an inch deep. In the centre of these cavities is a slight depression an inch in diameter. The edge is slightly convex, and about an inch wide. The distance from the outer cir cle of the concavity to the point where the bevelling ceases toward the circumference, is a little more than an inch. The entire stone is beautifully polished. The regularity with which these relics are fashioned, challenges our admiration. Upon, some of them, the workman of the present day, with all his modern im plements and. mechanical skill, could not improve. Adair tells us that these hurling-stones were, from "time immemorial rubbed smooth on the rocks, and with prodigious labour," 1 and Lafitau says that a !N"ortli American Indian sometimes spent his life iii making a
1 "History of the American radians," etc., p. 402. London, 1T73.
DISCOID AJj STOKES.
351
stone tomahawk, and that without finishing it." Festination proceeds from the devil, is an aphorism with which the Indians offered quarrel neither in theory nor in practice. "With them, time was of no conse quence. It entered not as an element into their daily calculations. Consecutive labor formed no part of their ordinary occupations. Consequently, the matter on hand was readily postponed in favor of sleep, amusement, or mere idleness. "We may well "believe, therefore, that these discoidal stones, so carefully formed of such hard material as quartz and even agate, and fashioned into their present symmetrical shapes simply by means of attrition with other stones, and perhaps, in some measure, through the agency of large wooden drills assisted by sharp sand and water, should, in their construction, have occupied "weeks, months, and even years of tedious, although desultory labor.
The general distribution of these stones shows that the game, for which they were manufactured, was in common esteem among the various Greorgia tribes. Most of them are iniperforate, although some have come under the writer's observation which are so thor oughly perforated that they are little more than rings of stone. In a few instances, the discoidal stones with marked cavities seem to have been put to a secondary use, and treated as mortars" in which hard substances were triturated. For pulverizing clay, and perhaps some mineral substances serviceable for paint, they would have answered "well. If our conjecture as to the primary use of these stones be correct, it will readily be perceived that in bowling them the outer edge only would come in contact "with the surface of
' " llceurs dcs Sauvages Amtriquams," tome ii., p. 110. Paris, 1724.
3 O 2-
ANTIQUITIES OF THE SOUTHERN INDIANS.
the ground. The constant pressure of the thumb against the centre on one side, and of the fingers against the middle of the other side (the forefinger adapting itself to the curve of the periphery), would have exerted a decided tendency to keep the interior of the cavities smooth.
When, therefore, we perceive in the bottoms of these saucer-shaped cavities unmistakable traces of abrasion, we are persuaded that such discoidal stones have been diverted, from the use they were orginally designed to subserve. Such instances are, however, rare; and, in all likelihood, afford evidence not only of desiietude, but also of perversion, at the hands of mod ern Indians. Occasionally we have observed a solid or lenticular-shaped stone which gave indications of its having been employed at some later period as a grind ing or'mealing stone,
.Although it has been suggested that a similar game was practised among the ancient inhabitants of Cornwall, and perhaps at Nidau-Steinberg and Ebersberg,' this Chuugke game, by what name soever called, and however variant the rules which governed its de tails, -was essentially an American amusement, com manding almost universal favor at the hands of the North American tribes.
In the plains and upon the mountains of Chill may be seen numbers of flat circular stones, five or six inches in diameter, made either of granite or porphyry, and with a hole drilled through their middle. While Molina supposes them to have been the clubs or maces of the ancient Chilians, it is not impossible that some of them may have been used as gamiug-stones." We have
1 Killer's " Lnke Dwelling?," etc., pp. 135, 136. London, 1866. Tlates xxxviii. anil Ixxxix.
- Stcvens' "Flint Chips," p. 505. London, 1870.
DISCOID AL STONES.
353
seen how generally distributed these i-elics are through the Southern States. Several varieties, obtained by the Rev. Dr. George Howc just after a freshet in the Congaree River, which, overflowing its banks, laid bare an ancient Indian burial-ground not far from the city of Columbia, in the State of South Carolina, are represented in the plate which faces page 178 of the sixth volume of Schoolcrafb's " Archives of Aborigi nal Knowledge*" * Mr. Schoolcraft 3 states that " the numerous discoidal stones that are found in the tumuli and at the sites of ancient occupancy in the Mississippi Valley, serve to denote that this amusement was prac tised among the earlier tribes of that valley at the mound period. These antique quoits are made with great labor and still from very hard and heavy pieces of stone. They are generally exact disks, of a concave surface, with an orifice in the centre, and a broad rim." He expresses the opinion that the object of hurling these perforated specimens was " manifestly to cover an upright pin or peg driven into the ground." This. with all due respect, "vve question. The weight of authority inclines us to the belief that these stones were rolled, not pitched.
Messrs. Squier and Davis found these disks in the mounds of the "West; and, in their " Ancient Monu ments of the Mississippi Valley/1 have made us familiar with their characteristic types. 3 They are related, say they, to a very numerous class of relics scattered over the surface from the valley of the Ohio to Peru, com posed of granite, porphyry, greenstone, jasper, quartz,
1 Philadelphia, 1860. 2 " Archives of Aboriginal Knowledge," etc., vol. u, p. 87, plate 23. Philadel phia, I860. 3 " Smithsonmn Contributions to Knowledge, 1 ' vol. i., p. 221, et seg., Fig. 121. Washington, 1848.
23
351
ANTIQUITIES OF THE SOUTHEI-MST INDIANS.
etc. Although the opinion of these gentlemen is enti tled to the highest consideration, in the light of subse quent investigations we feel constrained to differ from them "when they advance the idea that discoidal stones are " of comparatively modern origin." They are old, very ancient, if we understand the record of the mounds in which they have been, found, and rightly interpret the antiquity of the relics with which some of them are associated. The game, of which they are the symbols--conceived in a distant past and maintained through the intermediate centuries--was still popular within historic times. To the accounts already given, it may prove interesting to add a few other notices:
" The Rev. J. B. Finley (distinguished for his zealous efforts in Christianizing the Indian tribes of Ohio)," say Messrs. Squier and Davia,1 " states that among the tribes with which lie was acquainted, stones identical "with those above described, were much used in a popu lar game resembling the modern game of ' ten pins.' The form of the stones suggests the manner in which they were held and thrown, or rather rolled. The concave sides received the thumb and second iing-er, the forefinger clasping the periphery."
Mr. Breckenridge * mentions a game popular among the Iliccarees which was played "with a ring of stone; and Lewis and Clarke assert that a similar amusement was indulged in by the Mandans. The javelin-game among the Pawnees was probably but a modification of this ancient sport.*
To the Abbe Em Domenech * -we are indebted for
1 " Smithsoniaii Contributions to Knowledge," vol. i., p. 223, note. 2 "Views of Louisiana," p. 256, quoted by Squier and Davis. > " Travels in Xortli America," Hon. C. A. Murray, vol. i., p. 321. Morgan's "League of tlie Iroquois, pp. 299-302. 4 " Seven Years' Residence iu the Great Deserts of North America," vol. ii. . 197. London, I860.
,
I I *
j ^ i
f ... p [i
\
j i
i j' i
GAME OF SPEAK AKD BIWG.
355
tile following : " Their game of Spear and Ring is extreinely curious and difficult. The players are divided into two camps, for Indians are fond of collective par ties in "vvhicli are many conquerors, and. consequently many conquered. The stakes and bets are deposited in the care of an old man ; then a hard smooth ground, without vegetation of any kind, is chosen, in the middie of which, is placed perpendicularly a stone ring of about three inches diameter. When all is prepared, tile players (armed with spears six or seven feet long, furnished with small shields a little apart from each other, sometimes with bits of leather) rush forward, two at a time, one from each camp ; they stoop so as to place their spears on a horizontal level with the ring, so that they may pass through it--the great test of skill being to succeed without upsetting it. Each small shield or bit of leather that passes through, counts for a point : the victory remains to the player who has most points, or he who upsets the ring at the last hit.
" Some Indians render the game still more difficult by playing it as follows : One of the players takes the ring in his hand and sends it rolling, with all his strength, as far as possible on the prepared ground; his adversary "who is by his side, starts full speed after it to stop it, so as to string it on his spear as far as the last little shield.
" The Mojaves had a game so similar to the above that to avoid repetition it need only bo mentioned. The Natchez favorite pastime was very like the spear game, except that it required more strength and ad dress. Only two men conld play at a time. One threw with all his strength, and as far as possible, a long stick of the shape of a bat, and before it came to the
356
ANTIQUITIES OF THE SOUTHERN INDIANS.
ground, rolled a huge circular stone In the same direc tion. His adversary then threw a stick like the first, and he whose bat came nearest the stone gained a point and the right to launch the stone in liis turn: which was a great advantage, as from the impulse lie gave it a player was able to guess about how far the stone would roll."
Speaking of the Mandans, Catlin' says: " The games and amusements of these people are in most respects like those of the other tribes, consisting of ball plays --game of the moccasin, of the platter--feats of arch ery--horse-racing, etc.; and they have yet another, which may be said to be their favorite amusement, and unknown to the other tribes about them--the game of Tchung-kee, a beautiful athletic exercise, which they seem to be almost unceasingly practising whilst the weather is fair, and they have nothing else of moment to demand their attention. This game is decidedly their favorite amusement, and is played near to the village on a pavement of clay, which has been used for that purpose until it has become as smooth and hard as a floor. For this game two champions form their respective parties by choosing alternately the most famous players until their requisite numbers are made up. Their bettings are then made, and their stakes are held by some of the chiefs or others present. The play commences (plate 59) with two (one from each party), who start off upon a trot, abreast of each other, and one of them rolls in advance of them, on the pavement, a little ring of two or three inches ill diameter, cut out of stone ; and each one follows it up with his ' tchung-kee' (a stick of six feet in length,
1 " Illustrations of the Manners, Customs, and Conditions of the North Ameri can Indians," etc., seventh edition, vol. i., p. 132. London, 1848.
TOUTING--KEE GAME.
357
with little bits of leather projecting from its sides of an incli or more in length), "which he throws before him as he runs, sliding it along upon the ground after the ring, endeavoring to place it in such a position when it stops, that the 1'ing may fall upon it, and re ceive one of the little projections of leather through it, which counts for game one. or two", or four, accord ing to the position of the leather on which the ring is lodged. The last Avinner always has the rolling of the ring, and both start and throw the tchung-kee together; if either fails to receive the ring or to lie in a certain position, it is a forfeiture of the amount of the num ber he was nearest to, and he loses his throw ; when another steps into his place. This game is a very diffi cult one to describe, so as to give an exact idea of it, unless one can see it played--it is a game of great beauty and fine bodily exercise, and these people be come excessively fascinated with it; often gambling away every thing they possess, and even sometimes, when every thing else was gone, have been known to stake their liberty upon the issue of these games, offering themselves as slaves to their opponents in case they get beaten."
!No longer is this ancient game played either in Georgia or within the limits of conterminous States. Like the exercise of the discus in the heroic age, it has now become only a tradition--a shadowy memory from a nebulous past. The carefully-prepared areas over "which, from morning until night, the red athletes rushed hither and thither in the enthusiastic pursuit of this sport, at the expense of time and property and personal liberty, are entirely deserted now and rugged with the trunks and roots of huge forest-trees. The anointed poles and the swift hands which launched
358
ANTIQUITIES OF THE SOUTIIEBTf TSDIAWS.
them, have alike crumbled into nothingness. AVinners and losers, forgetting their profits and losses, the exulta tions and the disappointments of this exciting amuse ment, are themselves forgotten, and but little remains to remind us of the former existence and prevalence of this great game, characterized by singular dexterity, severe exercise and desperate ventures, save these discoidal stones, often so remarkable for their symmetry, and so expressive of the skill and labor expended in their manufacture.
CHAPTER XVI.
Stone Tubes.
THE tube, of which Fig. 1, Plate XXI., is a clever representation, was found in a small burial-mound in Burke County. It is made of serpentine, is thirteen inches and a quarter in length, and weighs three pounds and a half avoirdupois. The bore at one end is circular, and an inch and three-eighths in diam eter. At the other end the commencement of the aperture is elliptical in shape, the length of the major axis being an inch and three-eighths, and that of the minor one inch. At this end the exterior surface has been correspondingly flattened. The walls of the tube are about three-eighths of an inch in thickness. The perforation or bore extends longitudinally through the implement, but with diameters gradually diminishing from either end, until, at the point where they meet in the centre, the opening, which is there circular, is less than a quarter of an inch in diameter. In this tube the hollow has been compassed not by drilling but by gouging out or removing the interior particles of stone by means of some sharp-pointed instrument. Longi tudinal scars, caused by the operation, are still dis cernible on the inner surface. No circular strise can
3()0
ANTIQUITIES OF THE SOUTHERN INDIANS.
be perceived. The elevated ring around the central portion is rudely ornamented. with zigzag lines. ^Vith tlris exception the exterior surface is smooth, entirely plain, and well polished.
Similar tubes of steatite (Fig. 2, Plate XXI.), soapstone, and mica slate, have been found in other localities--all, however, smaller than the specimen we have just examined. Seven indies would express trie average length of such as have passed under the writer's observation. The bores of some of them are, at each end, of like diameter--say an incli and threequarters--gradually lessening as they approach the centre, where the width of the aperture does not mate rially vary from a quarter of an inch. The Avails are sometimes .less than a quarter of an inch in thickness. Upon the ornamentation of the ends and exterior sur face of these tubes, in some instances, considerable pains have been bestowed. Stone tubes not precisely similar in character, but bearing a general resemblance, have been described and figured by Mr. Schoolcraft,1 Messrs. Squier and Davis,2 and others. .An implement very like in its conformation is thus noticed by Mr. Haywood: 3 "About eighteen miles east from Rogersville in the county of Hawkins, in East Tennessee, was ploughed up a stone trumpet. It tapers on the out side from either end to the middle, and is there sur rounded by two rings of raised stone. The inside, at each end, is a hollow of an inch and a quarter in diam eter ; but at one end the orifice is not as large as at the other. Probably the sound is shrill and. sharp when blown from one end, and more full and sonorous when
STOIO: TUBES.
361
blown from the other. The hollow continues through out from the one end to the other, but in the middle, under the rings, it is not as wide as at the ends. It seems to have been made of hard soapstone; and, when blown through, makes a sound which may be heard perhaps two miles. It is very smooth on the outside, but rough within."
By some, Mr. Schoolcraft among the number, it has been suggested that these tubes "were telescopic de vices; and, in their construction, he pretends to trace an analogy to the tubular chambers used by the A.ztec and IMaya races in their astronomical observations. This notion we regard as fanciful; nor do we sympa thize in the belief of those who pronounce them musical instruments. In vain have "we endeavored to evolve a single sound beyond a dull, dead blast; and that in capable of transmitting itself to any practicable dis tance.
"We know that the Southern tribes were fond of music and dancing, and that their music was both vocal and instrumental. Aside, however, from their drums, tambours, rattle-gourds, and flutes made of the joint of a reed, or of the deer's tibia, they possessed no musical instruments worth the mention. 1 So far as present -recollection serves us, nowhere do we read of the use of stone trumpets, or any thing of the sort. It is entirely improbable that the Indians would have expended so much labor to such little purpose when the joint of a swamp-cane^ or a large conch, would have so readily, and so much better answered the de sired object. "We incline to the opinion that these were medicine or eupping-tubes.
1 See Bartram's " Travels through North and South Carolina, Georgia," etc., p. 502. London, 1Y92.
362
ANTIQUITIES OF THE SOUTHERN INDIANS.
Goreal * mentions that when the Florida Indians " are sick, they have not their veins opened, as is done elsewhere ; but they call for the Jaounas who are their priests and physicians. These suck that part of the body "which causes the patient most pain ; and they do it either -with the mouth or with a kind of shepherd'sflute (c1ialumeau~), having first made a small incision near some vein."
Cabega de Vaca 2 alludes to similar treatment of the sick: " The practitioner scarifies over the seat of pain, and then sucks about the wound. They make cauteries with fire, a remedy among them in high re pute, which I have tried on myself and found benefit from it. They afterward blow on the spot, and hav ing finished, the patient considers that he is relieved." Eibas, a century afterward, furnishes an account of this curative process : " The method of cure the possessed practitioners have, is sucking the part that aches ; if it be injured, bio-wing on it: which, for the effort and force, may be heard many steps off. . . . They give the sick to understand that the causes of their illness are the sticks, thorns and pebbles in their bodies, which they take out. This is false. They have the things in the mouth, or held craftily in the hand, and afterwards exhibit them as our tooth-pullers do teeth, on a string, as evidences of their professional skill." " After enumerating the cures by burning, smoking, scarifying, and sweating, Beverly states that the Vir ginia Indians sometimes made use of reeds for cauter-
MEDICINE--TUBES.
363
izing, which they heated over tlie fire "until they were on the eve of ignition, and then applied, upon a piece of thin wet leather, to the part affected.1
In his account of the aboriginal inhabitants of the Californian Peninsula, the German Jesuit missionary Ja cob Baegert, commenting upon the state of the medical art as it existed among the Indians of that region dur ing the second half of the last century, writes : " There are many impostors among them pretending to possess the power of curing diseases, and the ignorant Indians have so much faith in their art that they send for one or more of these scoundrels whenever they are indis posed. In treating a sick person, these jugglers, em ploy a small tube which they use for sucking or blow ing the patient for a while, making also various gri maces, and muttering something which they do not un derstand themselves, until, finally, after much hai'd breathing and panting, they show the patient a flint, or some other object previously hidden about their persons, pretending to have at last removed the real cause of the disorder." a
Venegas * confirms the observation of Baegert with regard to the use of stone tubes by the medicine-men of the California Indians: " One mode was very re markable, and the good effect it sometimes produced heightened the reputation of the physician. They ap plied to the suffering part of the patient's body the cluicuaco, or a tube formed out of a very hard black stone ; and through this they sometimes sucked, and other times blew, but both as hard as they were able.
" History and Present State of Virginia, book in., cliap. is p. 49. London,
" Natural and Civil History of California," vol. i., p. 9Y. London, 1759.
364
ANTIQUITIES OF THE SOUTHERN INDIANS.
supposing that thus the disease was either exhaled or dispersed. Sometimes the tube was filled with dinarTon or wild tobacco lighted, and here they either sucked in or blew down the smoke, according to the physician's direction ; and this powerful caustick some times, without any other remedy, has been known en tirely to remove the disorder."
These authorities confirm our impression that tubes --like those we have been considering--were medici nal in their uses, and materially assisted the primitive physician--at once quack and conjurer--in perform ing his wonderful cures. The flattened end appears adapted to the lips. This, and the small hole in the centre of the bore, facilitated both the blowing and the sucking process. By the circular opening at the larger end, the seat of pain, could have been conveniently covered. The weight of the instrument enhanced its efficiency, and rendered more facile its preservation in the .desired position. Wliile under treatment, Indian patients were compelled to assume more than a recum bent position. They were obliged to lie flat down, now on the back, and again on the stomach. If we go one step farther and suppose the cavity next to the flattened end filled with punk, dried tobacco-leaves, or some combustible material, the other end of the tube being firmly applied to the part affected, which had been previously scarified, we will perceive, when the contained substance was ignited, how readily this tube would have answered the purposes either of cauteriza tion or clipping. In the one case the particles of burn ing matter dropping through the central opening would, have blistered and burnt the diseased spot; while in the other, the active fire kindled in the upper portion of the tube--the ashes by a simple contrivance being
OETIAMENTAL TUBES.
365
prevented from falling through the narrow portion of the bore below--would have created and maintained during its existence a vacuum in the lower part of the tube, thus causing the blood to flow freely from the incisions made in the flesh.
Other tubes (Figs. 3, 4, 5, and 0, Plate XXI.) occur, which, apparently, were used as ornaments. These vary in length from two to three inches and a half, and in general conformation resemble triangular prisms, with convex sides, and angles slightly rounded. Steatite, talcose slate, and soapstone, were the custom ary materials employed in their construction. Perfo rated longitudinally, the average diameter of theii' bores may be stated at from three-eighths to one-half of an inch. Numbers of such relics have been found along the banks of the Savannah River above Augusta, and in other portions of the State. The exterior sur face is not infrequently ornamented with incised lines, curved, straight, and zigzag. Ordinarily, the holes were drilled, the circular striae being clearly defined, and the bore of equal diameter throughout the entire length of the article. Our impression is that they were worn as ornaments. In Fig. 125, of the " Ancient Monuments of the Mississippi Valley," ' are represented two varieties of stone tubes of this class.
CHAPTER XVII.
Stones for rounding Arrow-shafts.--Whetstones or Sharpeners.--Pierced Tablets. --Pendants.--Slung-sloiies.--Amulets.--Stone Plate.--Mica Mirrors.--Sculp tured Rocks.
AT a large Iranting-camp, which had been aban doned by the Indians, Captain Romans * noticed " some stones deeply marked by the savages -with some un couth marks, but most of them straight lines and crossed." He conjectured that they had been used for grinding awls. The only means the natives possessed of restoring an edge to their -worn and blunted axes and other stone implements was, by rubbing them against "whetstones. Hence "we frequently meet with irregularly-shaped stones, grooved and scarred by this process. It will be remembered that the hard canes of the Southern swamps supplied the red hunters of this region with convenient and abundant store of arrows. The material was most suitable for this pur pose, combining, as it did, requisite size, durability and lightness. No labor -was necessary in shaping the arrow, save such as -was expended in removing the exterior sheath, in smoothing the joints, in straighten ing and in cutting the reed off at the desired length.
1 " Concise Natural History of East and West Florida," p. 827. New York, 1775.
WHET STONES.----PIERCED TABLETS.
367
the green cane was subjected to a certain degree of heat, its natural moisture was readily expelled, and the reed easily freed from any irregularities. If made, while heated, to assume a direct line, it M'ould not de viate therefrom when cold and dry. In order to facili tate this straightening and polishing of the arrow, it was, while hot, passed through grooves made ill sand stone or in some other coarse-grained stone. These grooves are generally carefully made in direct lines, are even in their diameters, and frequently intersect eacli other at right angles, thereby presenting the ap pearance of crosses. An example is represented in Fig. 1, Plate XXII.
It will be perceived that the heated cane arrow, when pressed and rubbed in these grooves, would not only be freed from all irregularity of surface, but would .also be compelled to assume a direct line. Such a con trivance equally facilitated the manufacture and polish ing of wooden arrows--the rough surface of the stone acting as a file in reducing the shaft to the desired size and rouiidness. These STORES FOB ROTJNDING- ARROWSIIAFTS are readily distinguished from the ordinary WHETSTONES, so generally employed for sharpening the edges of axes and other cutting implements. (Fig. 7, Plate XXII.)
PIERCED TABLETS.--Various as the fancies of the makers are the shapes of these relics.
The illustrations prepared by Messrs. Squier and Davis of the " gorgets " of the Mississippi Valley ' aptly represent most of those generally found in the mounds, relic-beds, and fields of Georgia and her sister States. Many of them were made of a beautiful slate, with
3(jtf
ANTIQUITIES OF THE SOUTIfEISN" INDIAIVS.
from two to five perforations, and are so thin and deli cately constructed, that they could have served only the purposes of ornament. Pierced tablets of this class are lozenge-shaped, oval, parallelogrammic, or fash ioned after the similitude of the lid of a coffin. Oth ers, again, are cruciform and star-shaped. In fine, their forms are varied, and in many instances quite fanciful. Many possess only a single perforation in the upper end or near the middle. Most, however, have two holes drilled, about an inch apart and through the cen tral portion. They vary in length from three to eight inches, in width from three-quarters of an inch to three inches, and in thickness from the sixteenth to a quarter of an inch. The edges are frequently ornamented with notches, and the l>road surfaces are sometimes covered with incised lines. In all instances of this ornamental class it will be observed that the perforations are uni form, generally varying from the eighth to a quarter of an inch in diameter. Objects of this fragile descrip tion were, we think, intended as ornaments, and were suspended from the neck or fastened to some conspicu ous part of the vestment.
There is another variety, however, so much in ore substantial in its character that it seems to have been designed for practical, use. Relics of this class are made of serpentine, of greenstone, and of hard slate. Even jasper has been employed in their manufacture. They are thick and durable. It is not uncommon to meet with one of them fully an inch in thickness, although most of them do not attain that dimension by a half. They contain in most instances only two perforations, located in the central part of the imple ment, and about an inch or an inch and a quarter distant from each other. These perforations--unlike those ob-
PIEECED TABLETS.
36^
served in the case of tlie ornamental piercad tablets-- are conical in form. The drilling is never done from both sides, but only from that side where the aperture is largest. The shape of the perforation is evidently not accidental, because a uniformity exists. The aper ture on one side is about twice as wide as it is on the other. Such is the general rule in the case of the thicker "gorgets," and hence it has been suggested, that implements of this class were employed by the Indians in the manufacture of their bow-strings. The material used for this purpose, it is believed, could be readily pressed through the wider opening and then drawn so as to make it conform to an even size. Several thong's thus passed through the two or three apertures in the same gorget, when drawn to the required length, all being of the same size, could conveniently have teen twisted into one common, strong cord. Of this vari ety we figure two typical specimens (Figs. 2 and 3, Plate XXII.).
In response to a letter from Prof. Charles Ran, Mr. Catlin writes : " AVith regard to the tablets of which you speak, I have seen several, but the holes were much larger than those you describe. Those which I have seen were used by the Indians for grooving1 the shafts of their arrows. .All arrows of the primitive In dians are found with three grooves extending from the arrow's shoulder at the fluke, to the feathers, aud con ducting the air between them so as to give them steadiness. These grooves, on close examination, are found to be indented by pressure, aud not iu any way cut out; and this pressiire is produced, while forcing the arrow softened by steaiu through a hole in the tablet, with the incisor of a bear set firmly in a handle and projecting over the rim of the hole, as the arrow-
370
ANTIQUITIES OF THE SOUTHERN INDIANS.
shaft is forced downward through the tablet, getting compactness and on the surface and in the groove a smoothness which no cutting1, filing, or scraping can produce. It wonld be useless to pass the bowstring through the tablet, for the evenness and the hardness of the strings are produced much more easily and effectually by rolling them, as they do, between two flat stones whilst saturated "with heated glue."
It thus appears that this extensive and venerable observer discountenances the idea that these perforated tablets were employed in the manufacture of bow strings, thongs, and cords. Since, through the kind ness of Prof. Rau, we have been made acquainted with these remarks of Mr. Catlin, we have carefully examined the tablets in our possession and have failed to note any impression produced upon the edges of the perfora tions by the pressure of any thing like the incisor of a bear. Remembering the rather soft material from which many of these tablets were made, we might rea sonably anticipate the presence of some abrasion in the perforations if they were sttbjected to a constant use such as has been suggested1.
PENDANTS.--The typical forms indicated by Messrs. Squier and Davis in the " Ancient Monuments of the Mississippi Valley," 1 have their counterparts among the relics of the Southern Indians. The pear-shaped pen dant, "with groove around the upper end, is by no means infrequent. Brown hematite, greenstone, quartz, and a variegated jasper, were the materials selected for its manufacture. Of this variety some are so delicate, so beautiful, and so carefully polished, that they seem to have been designed for nose and. ear ornaments.2
1 Page 235. "Adair ("History of the American Indians," pp. 170, 171. London, 1776) alludes to the use of course diamonds and bits of stone fastened with deer's sinew
PEKDAKT3.
81 1
IVIost of them, however, are so heavy that they could not well have answered such a purpose. May not the larger sorts have been employed--after the fashion of the modern bobbin--in twisting bow-strings, plaiting belts, and in weaving variotts articles for personal decoration ?
From the peculiar shape and the careless manner in which many--found in the relic-beds along the river-banks--have been fashioned, it seems probable that they were intended as fishing-plummets. In their construction soapstone was the favorite material used. Often triangular in shape, sometimes they ap pear in the form of a double conoid, with a groove around the middle. These are usually so much lighter than the net-sinkers, and differ so essentially from them in figure, that they need not be confounded "with them. Specimens of this class often resemble num ber 5, figure 132, "Ancient Monuments of the Missis sippi Valley," althoitgh, instead of being notched at the upper end, many have a groove around the middle ; while others--made of soapstone or slate--are trav ersed by longitudinal as well as transverse grooves.
Nearly allied to the pear-shaped pendant is an in strument which, wThen first observed by the "writer, he regarded as a pendant whose upper end had been broken off and then flattened by attrition. Other relics, however, identical in shape, proved that the form was designed and not accidental. Carver ob served among the Indians living westward of the Mis sissippi River a warlike implement consisting of a stone of middling size, curiously wrought and fastened
:ar, and the maceaseene; and Lawson (" History of reprint, 1860) declares that some of the Indians wore
372
ANTIQUITIES OF THE SOUTHERN INDIANS.
by a string, about a yard and a half long, to the right arm a little above the elbow. Such stones the natives carried ill their hands until they approached their ene mies, when, riding at full speed, they swung them with great dexterity and never failed of doing execu tion.'
Among the Shoshoiiee Indians, Lewis and Clarke noticed an instrument consisting of a handle about the size of a whip-handle, made of wood, twenty-two inches long and covered with leather. At one end was a thong two inches in length, to which was at tached a stone weighing two pounds, enclosed in a leather cover. At the other end was a loop by means of which the implement was secured to the wrist. With this weapon they struck a powerful blow. It may be that the pear-shaped stones, now under exam ination, were made for some such purpose, and that they were carried and handled very much as slungshots are used in the present day. Those in our pos session are about as large as a turkey-egg, closely re sembling it ill shape, save that the pointed end has been cut off at right angles. {See Fig. 4, Plate XXII.) They have not the clearly-defined necks possessed by the relics delineated in Fig. 117, of the " Ancient Monu ments of the Mississippi Valley." * Stones, for throw ing by hand and perhaps by means of a sling, occtir fre quently. They are commonly round' or ovoidal, and appear to have been gathered from the beds of streams --or rudely fashioned from soapstone.
AMULETS.--Fig. o, Plate XXII., represents a class of objects frequently found in Ohio and in other portions of iSTorth America, but seldom seen in that part of the
i " Travels," etc., pp. 294, 295. London, ITiS. ! FMge 219.
AMULETS.----STO2CE PLATES.
373
country once occupied by the Southern tribes. Gen erally made of a greenish striped slate, they are, in most instances, designed to represent a bird. Their use is not well understood, but it is probable that they possessed some conventional significance and importance in con nection with the religious ideas of the Indians. Three of these strange articles are figured on page 239 of the " Ancient Monuments of the Mississippi Valley," and their varying forms are richly shown in the valuable collection of the Smithsonian Institution. It has been suggested by some that they were used for husking Indian-corn. This idea we regard as entirely fanciful. It appears much more probable that they were es teemed and worn as charms, as badges of distinction, or as religious tokens.
STONE PLATE.--So far as our information extends, this relic (Fig. 6, Plate XXII.) was the first of its kind fotincl within the limits of the United States. It was ploughed up in 1859, on the lower terrace of the large temple-mound on the Etowah Hiver, upon the planta tion of Colonel Lewis Tumliu, near Cartersville. This interesting locality has proved the thesaurus of more valuable and curious aboriginal remains than any other spot in Georgia. To the companionship of the terraced mound, the stone idols, idol-pipes, simulacra of various sorts, fish-preserves, gold and pearl beads, shell orna ments, ising-glass mirrors and sundry beautiful imple ments of diorite, hornblend, jasper and flint, may now be added this stone plate, circular in form, eleven inches and a half in diameter, an inch and a quarter in thickness, and weighing nearly seven pounds. It is made of a close-grained, sea-green slate, and bears upon its surface the stains of centuries. Between, the rim, which is scalloped, and the central portion, are two circular de-
374
ANTIQUITIES OF TIIK SOCITIIEKK INDIANS.
pressed rings, running parallel with the circumference and incised to the depth of the tenth of an inch. The central portion, or basin, is hollowed out to the depth of rather more than the eighth of an inch. This cir cular basin, nearly eight inches in diameter, is sur rounded, by a margin or rim, a little less than two inches in width, traversed by the incised rings and bevelled from the centre toward the edge. The lower surface or bottom of the plate is flat, bevelled upward, however, as it approaches the scalloped edge, which is not more than a quarter of an inch in thickness.
Two stone plates, similar ill material, size, and general configuration, were unearthed in this locality. "Within the past eighteen months, two relics of this class, but less elaborate in their construction and smaller, were obtained from a mound on the Black "Warrior Kivcr in the State of Alabama. They now form a part of the collection of the Smithsonian Insti tution at Washington.
The use of these plates from the Etowah Valley may, we think, be conjectured -with at least some de gree of probability. It is not likely that they were employed for domestic or culinary purposes. Their weight, rarity, the care evidenced in theiv construction, and the amount of time and labor necessarily expended ill their manufacture, forbid the belief that they were intended as ordinary dishes from which the daily meal was to be eaten, and suggest the impression that they were designed to fulfil a more unusual and important office. The common vessels from which the natives of this region ate their prepared food were bowls and pans fashioned of wood and baked clay, calabashes, pieces of bark, and large shells. Flat platters, made of ail admixture of clay and pounded shells, well
OlfFEBIKGS TO IDOLS.
875
I
kneaded and burnt, were rdiiiai"i3y employed for
baking corn-cakes and frying meat; but it does not
anywhere appear that ornamented stone plates were in
'
general use.
It will be remembered, that at some remote period
idol-worship existed among many of the Southern
r
tribes. The religious duty of olfering fruits and food
to these lesser deities "was not neglected. "We are told
M\
that the "Virginia Indians represented their inferior
.
gods by the forms of men, calling such images Kewa-
I
sowolc. Tliese they placed in temples, ami in their
V,
presence worshipped, prayed, sung, and made repeated
I,
offerings. 1 Tlio stone column which Ribault placed upon a
i'
mound to mark the limit of tlie French empire ill the
I \ New World, was by-the Florida Indians elevated into
f
the dignity of a superior being. From top to bottom
it was encircled with flowers and the brandies of
choicest trees, while at its base were constantly ex
posed offerings of fruits, corn, favorite roots, bows and
t
arrows, and earthen vessels filled with perfumed oils."
TVhile the sun, as the most potent representative
of an unseen yet acknowledged divinity with supreme
*
powers, formed the chief object of religious worship
among the ancient tribes who peopled the Etowah
>;
Valley, there existed, nevertheless, images which, per
haps at the instance of designing priests and conjurers,
I
claimed the devotion of the masses. The precise posi-
|
tion assigned to them in the theogony of that rude age
I
does not fully appear; and yet, from, out the depths of
|
that dark period, comes light enough to reveal the fact
that these id.ols--subordinate though they were to the
1 Himot's " Virginia," p. 26. Priinooforti aJ Moemirn. Do Dry, anno 1590. 2 Plato viii. of the "Brevis Narratio."
376
ANTIQUITIES OF TIIE SOTJTIIEIOT
Great Spirit arid to the celestial luminaries--were still invested with attributes and influences which it be hooved weaker man to stand in awe of and to propi tiate.
\Ve incline to the opinion that these stone plates were designed for sacred uses, and that in them was exposed the food offered to the Dii Minores of those peoples who, antedating the modern Indians--dwellers here at the dawn of the historic period--erected the la,rge temple-mound in honor of that great (rod who mingled not with men, and before whose flaming min ister--the sun--they prostrated themselves in blind yet profound adoration.
MICA MEJIBKANACEA.--Large plates of ising-glass ave frequently found in the sepulchral tumuli of Geor gia, associated with articles of use and ornament, the property of the dead at the period of the inhumation. The largest which has fallen under the observation of the writer is elliptical in form, measuring thirteen inches in length, ten inches in width, and nearly half an inch in thickness. Usually, however, these mirrors --for such it appears proper to regard them--are much smaller. The customary size may be expressed by seven inches in length and five inches in width. Often elliptical, they are sometimes square or parallelogrammic in shape, and at other times quite irregular in their outlines. Being thick, and readily reflecting the opposed image, they answered tolerably well the pur poses of looking-glasses. "We are not aware that any specimens have been found backed by copper plates. 'If originally enclosed in frames, these were fashioned of such perishable materials that they long since crumbled, into nothingness, leaving no traces of their former pres ence or attachment. The frequent occurrence of these
JIIKKOKS OF MICA MEMBBAS'ACEA.
377
ising-glass mirrors, not only in the ancient graves and mounds font also upon the sites of old Indian villages and in relic-foeds, attests the fact of their general vise among the aborigines. Through their assistance, the process of personal decoration, of painting, and of tattooing, was materially facilitated ; and it is not improb able that they formed a source of special delight to many of the softer sex, who even in that rude age were not ignorant of their personal charms, or indifferent to snch artificial aids as might tend to enhance their focanty and attractions in the eyes of their savage admirers.
Occasionally a hole drilled through the lower edge of the plate--elongated after the fashion of a handle-- assures ns that the mirror "was sometimes suspended for convenient use, and that it was thus rendered more apt for facile transportation.
ScuxpTr/jrKD ROCKS.--In Forsyth County, Georgia, is a carved or incised foowlder of fine-grained granite, about nine feet long, four feet six inches high, and three feet foroad at its widest point. The figures are cut in the bowlder from one-half to three-quarters of an inch
deep. {See illustrations, p. 378.) A-S yet no interpretation of these figures has been
offered, nor is it known by whom or for what purpose they were made. It is generally believed, however, that they are the work of the Cherokces. On the eastern end of the bowlder, running vertically, is a line of dots, like drill-holes, eighteen in number, connected
by an incised line. Upon the Enchanted Mountain in Union County,
cut in plutonic rock, are the tracks of men, women, children, deer, bears, bisons, turkeys and terrapins, and the outlines of a snake, of two deer, and of a human hand. These sculptures--so far as they have been as-
878
ANTIQUITIES OF THE SOUTHEBK INDIAWS.
certained and counted--number one hundred and thir ty-six. The most extravagant among them is that known as the footprint of the " Great Warrior." It
South Side of the above.
measures eighte3ii inches in length, and has six toes. The other human tracks and those of the animals are delineated with commendable fidelity. One track-- which has been indifferently described as that of a horse and a buffalo--is seventeen inches long. These impressions are- artificial, and were scraped or chiselled out of the rock apparently with the aid of cutting im plements of flint. The accuracy and skill displayed in the construction of some of them challenge admi ration, while others are clumsily and rudely made. Most of them present the appearance of the natural tread of the animal in plastic clay. Twenty-six of these sculptures represent impressions of human feet-- varying in length from four to seventeen inches--all
INTAGLIOS.----EOCK--WRITHE G>.
379
of tliem bare, save one, which. was covered with a moc casin.'
These intaglios closely resemble those described by Mr. Ward as existing upon the upheaved slabs of coarse carboniferotis grit, in Belmont County, Ohio, near the town of Barnesville.2
Among the mountains which fence in the upper portion of Georgia, in several localities, may still be seen, carved in rock, similar intaglios and rude repre sentations of the sun, the human form and hand, the bow and arrow, the canoe, and various circles and ir regular figTires which, at the present day, seem almost meaningless exhibitions of the fancies of those by whom they were traced.
Intended, doubtless--especially when associated in groups--to perpetuate the recollection of some memo rable event, the histories which they chronicled and the traditions they were designed to transmit, have, like the peoples who formed them, quite faded from the memory of succeeding generations. As yet we have seen nothing of this sort which rises above the dignity of rude picture-writing, such as at later periods has been, in a more ephemeral way, practised by the modern Indians in commemoration of an engagement, in adoration of the sun, in token of an alliance, in ex planation of some marked occurrence, aiid in imitation of some well-known natural object. We search in vain for alphabet, lettered shaft, phonetic sign or digit, liutle representations all, they do but feebly shadow forth the earliest efforts at physical expression of com mon incidents, the most primitive attempts at com-
380
ANTIQUITIES OP THE SOUTHERN INDIAKS.
municating thoughts and perpetuating the recollec tion of events "by visible shapes. A few symbols there were, which possessed and retained an acknowledged significance, recognized by all.
Tlio labors of tlie aborigines upon the summit of Stone Mountain, so far as we have been able to ex amine them, seem to have been directed not, as has been surmised, to the cutting of hieroglyphics and the fashioning of curious figures in the granite, but to the preparation of little ditches or trenches for the protec tion of their fires. The mountain being entirely bald and consisting of hard granite, during a storm, the rain --unabsorbed by the rock--woxild fiow freely clown the sides and soon smother the fires kindled upon its surface. In order to avoid tills inconvenience, the In dians resorted to the expedient of cutting in the rock circular arid horseshoe-shaped troughs or trenches, which would catch the rain-water in its descent and divert it from the interior spaces upon which their fires were burning. A marked similarity exists in the sizes and shapes of these fireplaces, which are scattered in considerable numbers upon the summit and western slope of the mountain. They are generally from three to four feet in diameter, and are circular, semicircular, and elliptical in form. The incised trenches or ditches surrounding them are from four to seven inches in width, and. from two to three inches deep. In. the centre of almost every hearth is a fissure in the rock, which materially aided in preserving the fire. Of these fissTires the natives availed themselves in the location of their fireplaces; and, in some instances, at no little labor enlarged them and formed adjacent ele vations in the rock, as convenient resting-places for the earthen vessels in which they cooked their food.
SCTILPTUBE.
381
We have thus, in our opinion, a simple explanation of tie practical use of these incised Hues, artificial eleva tions, and trenches cut in the rock, which have been, supposed by many to possess a hidden and mysterious "significance.
The delightful temperature of this mountain--a stupendous pyramid rising in austere and solitary grandeur above the plain--during the summer at tracted the natives. That it was a favorite resort of the primitive peoples who in former centuries occu pied this region, is attested by the presence of these laboriously-constructed fireplaces, and by mortars, per manent in their character and hollowed out of the rock. This impression is confirmed by the traces of a defen sive rock-wall which at one period fortified the entire crest of the mountain.
It is above the wall, and around the summit of Stone Mountain, that the indications of long-continued occupation by the red-men are most abundant.
Returning from this digression, we would remark that in the stone images, idol, bird, and animal-shaped pipes, and in the large ornamented shell-gorgets, -we trace more emphatically than in any thing else the progress made by the Southern Indians in the art of sculpture. As the distinguishing peculiarities of these relics will, however, be considered in a subsequent part of this work, it is unnecessary to anticipate what will be said on the subject. These vai'ious devices and. imitations in stone and shell, while they often, ex hibit no little skill and ingenuity, are, after all, but rude expressions of the taste of the untaught, and fall far short of what may properly be considered works of art. Few and feeble were the attempts to transmit important memories by means of enduring physical
3S2
ANTIQUITIES OF THE SOUTHERN INDIANS.
signs and pictorial illustrations. Upon the rock-walls which guard the confines of Northern Georgia, we look in vain for any monumental traces of the liistory of trie tribes wlio were native here long before the advent of the European. Among the relies intermingled with the soil upon which they dwelt, we search fruitlessly for a single tablet whereon were engraven their laws or the names of their kings, priests, and warriors. They lived and died, peoples without letters, and the Muse of History finds scarce an epitaph for their tombs. Trusting to the trembling" voice of the aged warrior for a record of the brave deeds of their ancestors, and committing to the memory of the younger chieftain the story of their present achievements, they gave to the passing wind the spoken word, but carved not a line and reared not a column in commemoration either of the past or the present. History is voiceless where the use of iron and of letters is unknown. Under such circumstances, the most wTe can hope to discover is comprehended in vague traditions, and in the silent teachings of monuments and relics which have escaped the destructive influences of time. We compare, we conjecture, we speculate. The rest is darkness.
CHAPTER XVIII.
Pipes.--The Use of Tobacco.--Idol Pipes.--Calumets.--Comra
THE pipes of the JVorfch American Indians possess an importance, both traditional and historic, which, elevating them above the category of ordinary relics, claims for them a moral, religious, and political value, which must be duly appreciated in forming a suitable estimate of their office and in comprehending the va rious purposes they were intended to answer. It is not only as the media through which the narcotic in fluences of tobacco were imparted, nor as articles of pride and ornament npon which the protracted labor and best still of the primitive artists were expended, nor yet as the acknowledged symbols of alternate peace and war that they are to be regarded. Combin ing all these, they rise higher and confess themselves, in their origin, the immediate gift of the Great Spirit by whom they were invested with certain prescribed sanctities, and, in their nses, fenced aboxvt with positive injunctions "whose non-observance entailed disaster and supreme displeasure. Devotional also in their nses, their agency was invoked in solemnities which brought the red nomad face to face with his Creator. Then the pnffs of smoke blown to the four quartei's of the
3S4
AKTIQTJITIES OF THE SOHTIIEBa" IMDIABS.
heavens were redolent of the petitions of the devotees. Taking the place of sacrifice, through them propitia tion was made to an angered divinity. In sunshine and. almndance they "were the ministers of gratitude-- the exponents of joy and thanksgiving. Among the primitive inhabitants of at least some of the Southern, regions, they were elevated to the dignity of idols be fore whose elaborately-carved forms of man, and. beast, and bird, the deluded fell down and worshipped. It is only among the ISTorth. American Indians that such peculiar historic interest attaches to the Pipe--only among the ancient peoples of this region that we lo cate customs, ceremonies, and traditions, at once most curious and U-niqvie.
Standing on the precipice of tho red-pipe-stone rock of Goteau des Prairies, the Great Spirit broke from it a fragment, and, by merely turning it in his hands, made out of it a huge pipe 'which, having smoked, lie proclaimed a symbol of peace among all his children, declaring this stone common property, ordering peace-pipes to be fashioned from it, and for bidding the war-club and the scalping-knife to be lifted near it.' Prof. Wilson" very justly remarks that, "in the Old ^Vorld the ideas connected with the tobaccopipe are prosaic enough. The chibouk may, at times, "be associated with the poetical reveries of the Oriental day-dreamer, and the hookah with pleasant fancies of the Anglo-Indian reposing in the shade of his bunga low ; Taut its sedvietive, antirpae mystery, and all its symbolic significance, pertain to the ]STew World."
Longfellow, accordingly, fitly opens his Song of with the institution of " the Peace-Pipe."
CALUMET, OH PEACE--PIPE.
885
Tli3 Master of Life descends on the mountains of the prairie, breaks a fragment from the red stone of the <]uarry, and fashioning it witli curious art into a pipetead, fills it with the bark of the red-willow, chafes the forest into flame with the tempest of his breath, and kindling it smokes the calumet as a signal to the nations. The tribes of the ancient aborigines gather, at the divine summons, from river, lake, and prairie, to listen to the warnings and promises with which the Great Spirit seeks to guide them. This august audience concluded, the warriors having buried their war-club?, smoke their nrst peace-pipe and depart:
" \Vhile the Master of Ufe, ascending
Throngh the opening cloud-curtains Through the door-ways of the lieavcii,
Vanished from before their faces,
In the einolve that Tolled around him,
The pnlcwana of the peace-pipe."
For toba<%o a divine origin is said to have been claimed by the American Indians, who regarded it as a direct gift from the Great Spirit, for their special en joyment. Indeed, according to Ilariot, they believed that the Great Spirit was himself addicted to the habit of smoking. The pipe, therefore, camo to be regarded as a sacred object, and smoking partook of the character of a moral if not a religious act.' The incense of tobacco was deemed pleasing to the Father of Life, and the ascending smoke was selected aa the most suitable medium of communication with the world of spirits. The ordinary pipe was the constant companion and the unfailing solace of the Indian. TTpon the war-path, while engaged in hunting, and
'SteTunB' "FMutGI)lpa,"p. 338. London,18TO.
3Sb
ANTIQUITIES OF THE SOUTHERN IN"DIAH"S.
amid tlie lazy hours of liis mde liome-life, it was ever near, ministering to his pleasure and comforting him under misfortunes. Its introduction was essential to a formal declaration of war, and to the conclusion of a treaty of peace. Alternate whiffs were then tantamount to a signing1 and sealing by the parties in interest. Most important was the office of the Calu met. No tribal organization, no solemn assembly was complete without it, and the ceremonies observed in its honor were impressive and conducted with the utmost care and regularity.
" This Calumet" says Father Hennepin,1 "is the most mysterious Thing in the ^Vorld among the Savages of the Continent of the Northern A merica ; for it is us'd in all their important Transactions: However, it is noth ing else but a large Tobacco-Pipe made of !Red, Blade or "White Marble : The Head is finely polish'd, and the Quill, which is commonly two foot and a half long,' is made of a pretty strong Reed or Cane, adorn'd with Feathers of all Colours, interlaced, with Locks of A\romen's Hair. They tie to it two Wings of the most cu rious Birds they iincl, which makes their Calumet not much unlike Afercury's Wand, or that Staff Ambas sadors did formerly carry when they went to treat of Peace. They sheath that [Reed into the neck of Birds they call Huars, which are as big as our Geese, and spotted with Black and White; or else of a sort of Ducks who make their Nests upon Trees, tho7 AVater be their ordinary Element, and whose Feathers are of many different Colours. However, every Nation adorns the Calumet as they think according to their own Oeniiis and the Birds they have in their Country.
" A Pipe such as I have describ'd it, is a Pass and
1 " A N"e\y Discovery," etc., chapter xxiv., pp. 93, 94. London, 1698.
THE CAMTMET.
887
safe Conduct amongst all the Allies of tlie Nation who has given it; and in all Embassies, the Ambassadors carry that Calumet as the Symbol of Peace, which is always respected; for the Savages are gener ally persuaded that a great Misfortune would befal 'em if they violated the Publick Faith of the Calumet. All their Enterprises, Declarations of War, or Conclu sion of Peace, as -well as all the rest of their Ceremo nies are Sealed, if I may be permitted to say so, with this Calumet. They fill that Pipe with the best To bacco they have, and then present it to those with whom they have concluded any great Affair, and smoak out of the same after them. I had certainly perish'd in my Voyage, had.it not been for this Calumet or Pipe."
In Father Dablou's " Relation of the Voyages, Dis coveries, and Death of Father James Marquette," we have an interesting description both of the calumet and of the dance celebrated in its honor. I adopt the translation of Mr. John Gilmary Shea: 1 "It now re mains for me to speak of the calumet, than which there is nothing among them more mysterious or more es teemed. ]VIeii do not pay to the crowns and sceptres of kings the honor they pay to it; it seems to be the god of peace and war, the arbiter of life and death. Carry it about you and show it, and you can march fearlessly amid enemies who even in the heat of battle lay down their arms when it is shown. Hence the Iliiiois gave me one to serve as my safeguard amid all the nations that I had to pass on my voyage. There is a calumet for peace and one for war, distin guished only by the color of the feathers with which
3 " Discovery arid Exploration of the Mississippi Vulley," etc., p. 34, et seg. New York, 1S52.
388
ANTIQUITIES OF THE SOTJTHT5EN INDIANS.
tliey are adorned, red "being the sign, of war. They use them also for settling disputes, strengthening alli ances, and speaking to strangers.' It is made of a polished red stone, like marble, so pierced that one end serves to hold the tobacco while the other is fastened on the stem, which is a stick two feet long, as thick as a common cane, and pierced in the middle ; it is orna mented with the head arid neck: of different birds of beautiful plumage ; they also add large feathers of red, green, and other colors, 'with which, it is all covered. They esteem it particularly because they regard it as the calumet- of the sun ; and, in fact, they present it to him. to smoke, when they wish to obtain calm, or rain, or fair weather. They scruple to bathe at the begin ning of summer, or to cat new fruits, till they have danced it. They do it thus:--The Calumet Dance, which is very famous among these Indians, is performed only for important matters, sometimes to strengthen, a peace or to assemble for some great "war; at other times for a public rejoicing1 ; sometimes they do this honor to a nation who is invited to be present; sometimes they use it to receive some important personage as if they wished to give him the entertainment of a ball or comedy. In winter the ceremony is performed in a cabin, in summer in the open fields. They select a place surrounded with trees, so as to be sheltered beneath their foliage against the heat of the sun. In the middle of the space they spread out a large partycolored mat of rushes; this serves as a carpet, on which to place with honor the god of the one "who gives the dance; for every one has his own god? or maniton? as they call it, which is a snake, a bird or something of the kind, which they have dreamed in their sleep, and in which they put all their trust for the success of their
THE CALUMET DASCB.
389
wars, fishing and hunts. ISTear this manitoxi, and at its right, they put the calumet in honor of which the feast is given, making around about it a kind of trophy, spread ing there the arms used by the warriors of these tribes, namely the war-club, bow, hatchet, quiver and arrows. Things being thus arranged, and the hour for dancing having arrived, those who are to sing, take the most honorable place under the foliage. They are the men and the women who have the finest voices, and who accord perfectly. The spectators then come and take their places around undei* the branches ; but each one on arriving must salute the manitou, which he does by inhaling the smoke, and then puffing it from his mouth upon it, as if offering incense. Each one goes first, and takes the calumet respectfully, and, support ing it with both hands, makes it dance in cadence, suiting himself to the air of the song ; he makes it go through various figures, sometimes showing it to the whole assembly by turning it from side to side.
" A.fter this, he who is to begin the dance appears in the midst of the assembly, and goes first; sometimes he presents it to the sun, as if he wished it to smoke; sometimes he inclines it to the earth ; and, at other times he spreads its wings as if for it to fly ; at other times, lie approaches it to the mouths of'the spectato2-s for them to smoke, the whole in cadence. This is the first scene of the ballet.
" The second consists in a combat, to the sound of a kind of drum, which succeeds the songs, or rather joins them, harmonizing quite well. The dancer beckons to some brave to come and take the arms on the mat, and challenges him to fight to the sound of the drums ; the other approaches, takes his bow and arrow, and begins a duel against ths dancar who has no defence but the
31)0
ANTIQUITIES OF THE SOUTHERN INDIANS.
calumet. This spectacle is very pleasing, especially as it is always clone in time, for one attacks, tlie otlier defends; one strikes, the otlier parries; one flies, the oth.es pursues ; then he who fled faces and puts his enemy to flight. This is all done so well, with meas ured steps and the regular sound of voices and drums, that it might pass for a very pretty opening of a ballet in France.
"The third scene consists of a speech delivered by the holder of the calumet, for, the combat being ended without bloodshed, he relates the battles he was in, the victories he has gained ; he names the nations, the places, the captives he has taken, and as a reward, he who presides at the dance presents him with a beauti ful beaver robe, or something else, "which he receives, and then he presents the calumet to another, who hands it to a third, and so to all the rest, till all hav ing clone their duty, the presiding chief presents the calumet itself to the nation, invited to this ceremony in token of the eternal peace which shall reign between the two tribes.
"The following is one of the songs which they are accustomed to sing ; they give it a certain expression, not easily represented by notes, yet in this all its grace consists:
" 'Ninaliani, niunliani, niriahani, Nanioiigo.'"
The calumet of peace was frequently adorned with the white feathers of the bald. e,agle. lie who bore it passed freely, and without fear of bodily harm, wher ever he pleased; because this pipe--held sacred by all the tribes----rendered the person of him who carried it --bo lie chief, ambassador, friend, enemy, or stranger--
CALUMETS.
801
inviolable. Red being the color of war, Loskiel' tells us tliat in making peace or settling alliances, the redcalumet was " daubed over with M~hite clay or chalk."
By the same author we are informed that if two Indian nations entered into a treaty of alliance, a pipe of peace was exchanged between them, which was then called the pipe of covenant. It was carefully preserved and generally lighted in council whenever any thing oc curred appertaining to the alliance." Then each mem ber' smoked a little out of it. This reminded them in the most impressive manner of the covenant and the time of its establishment. The greatest care was be stowed upon the construction and oriiarfientation of the stems of the calumets and medicine-pipes. ]STo incon siderable official dignity attached to the bearers of them, and their preservation was a matter of earnest solicitude. "When M. D'lberville sought his first inter view with the Indians of Florida, he was received by their chiefs smoking the calumet and singing the song of peace. M. Penicaut * thus describes the pipe used on this occasion: "The calumet is a stick-about a yard in length, or a hollow cane, ornamented with the feath ers of the paroquet, birds of prey, and of the eagle. These feathers, arranged around, the stick, resemble somewhat the fans used by French ladies. At the end of this stick is a pipe, to which the name of calu met is given."
From Father Charlevoix " we borrow the following
1 " History of the Mission of the United Brethren among the Indians in North America," part 1, p. 1515. London, 1794.
2 " History of the Mission of the United Brethren," etc., part 1, page 158. London, 1Y94.
3 " Historical Collections of Louisiana and Florida," etc., by B. i\ French. New Scries, p. 38. Xew York, 1869.
4 " Voyage to North America," etc., vol. i., pp. 130, 181. Dublin, 1768.
ANTIQUITIES OF THE SOUTHERN INDIANS.
sting account of the character, uses, and impor tance of tlie calumet among tlie North American In dians: "The Calumet is not less sacred among these People than the Necklaces of Porcelain ; if you be lieve them, it is derived from Heaven, for they say it is a Present which was made them by the Sun. It is morj in Use with the Nations of the South and "West, than those of the North and East, and it is oftener used for Peace than for "War. Calumet is a 2forman word, which signifies Jieed, and the Calumet of the Savages is properly the Tube of a Pipe; but they comprehend under this Name the Pipe also, as well as its Tube. In the Cabimet made for Ceremony, the Tube is very long, the Bowl of the Pipe is commonly made of a Kind of reddish Marble, very easy to work, and which is found in the Country of the -Ajouez, be yond the Jlfississippi. The Tube is of a light Wood painted of different Colours, and adorned with the Heads, Tails and Feathers of the finest Birds, which is in all Appearance merely for Ornament. The Custom is to smoke in the Calumet whou you accept it, and perhaps there is no Instance where the Agreement has "been violated which was made by this Acceptation. The Savages are at least persuaded that the Great Spirit would not leave such a Breach of Faith unpun ished. If in the midst of a Battle the Enemy presents a Calumet, it is allowable to refuse it, but if they re ceive it, they must instantly lay down their Arms. There are Calumets for every Kind of Treaty. In Trade, when they have agreed upon the Exchange, they present a Calumet to confirm it, which renders it in some Manner sacred. When it concerns "War, not only the Tube, but the Feathers also that adorn it are red. Sometimes they are only set on one Side ; and
CALUMETS.
393
they say that according to the Manner in which the Feathers are disposed, they immediately know what Nation it is that presents it; and whom they intend to attack. There is scarce any Koom to doubt but that the Savages in making those smoke in the Calumet with whom they would trade or treat intend to take the Sun for Witness, and in some Measure for a Guar antee of their Treaties; for they never fail to blow the Smoke towards the Planet." Overlooking or else disregarding the current tradition that the pipe was' the direct gift of the Great Spirit delivered at first with specific injunctions and to be used on all impor tant occasions with becoming solemnity, and always with the greatest good faith, our author is of opinion that the Indians " having found by Experience that the Smoke of their Tobacco draws Vapours from the Brain, makes the Head clearer, rouses the Spirits and makes us titter to treat of .Affairs," for this reason introduced its use into their councils ; and that " after having gravely deliberated and taken their Resolution they thought they could never find a Symbol fitter to put a Seal to their Determinations,.or any Pledge more capable of confirming the Execution of them than the Instrument which had so much Share in their Delibera tions. . . . To smoke in the same Pipe therefore in Token of Alliance is the same Thiiig as to drink in the same Cup, as has been practised at all Times by many Nations."
^Vhen Columbus was upon the coast of Cuba he beheld several of the natives going about with fire brands in their hands, and,. certain dried hei'bs which they rolled up in a leaf, and lighting one end put the other in their mouths and continued inhaling and puffing out the smoke. A roll of this kind they called
394
ANTIQUITIES OF THVE SOUTHERN INDIANS.
a tobacco, a name since transferred to tlie plant of which the rolls were made. The Spaniards, although prepared to meet with wonders, were struck with as tonishment at this singular and apparently nauseous indulgence. 1
In his chapter upon the nations and tongues of Florida, Cabeca de Vaca states that the natives every where " produce stupefaction with a smoke, aud for that they will give whatever they possess." 3 This al lusion--brief and vagne although it be--has reference to the use of tobacco, and assures us emphatically both of the fondness of the Indians for that "weed, and of the extravagant degree in which they subjected them selves to its narcotic influences.
One of the oldest references to Indian pipes is con tained in the " Brevis Narratio " of Lo Moyne de Mor gues. Plate xx. represents various methods of cur ing the sick. An Indian is seated, smoking a pipe. A woman offers him some tobacco-leaves. The text is as follows : " Quandam etiam planta habent cujus nomen exeidit, Brasiliani PETUM, Hispani TAPACO appellant: hujus folia probe siccata laxiori tnbuli parti imponunt, eorum incensorum fumum angustiore tubiili parte ori admota attrahunt tarn valide, ut per os & nares illis egrediatur, & eadena opera abnnde humores eliciat.'' This passage may be translated thus : " They have also a certain plant whose name I have forgotten--the Bra zilians call it PETIJM, the Spaniards TAPACO--whose well-dried leaves they place in the wider portion of a tube. Having ignited these, they apply the narrower part of the tube to the mouth aud draw out the fume
PIPES.----TOBACCO.
395
so vigorously that it escapes through tlie mouth, and nostrils, and tlius removes much humidity."
We have also, in the first volume of Do Bry, an other representation of an Indian pipe. It is not un like the short clay-pipe of the present day, so much in vogue among the Irish laborers. Beverly, in his " His tory of the Present State of Virginia," has reproduced this illustration. 1
The illustration which faces page 7 of the third book contains two representations of the pipe of peace ; and Father Hennepin, in the frontispiece to his " New Discovery," figures a naked Indian holding in. Ms hands the plumed calumet. Carver" has furnished us with a drawing of the pipe of peace.
In the early narratives smoking is alluded to rath er as a curative process or public ceremony, than as a matter of habit or enjoyment among the natives. The ignorance of the times and the novelty of the custom furnish plausible excuse for the mistake.
In his u Briefe and True Report of the New-found. Land of Virginia," ' Hariot thus quaintly describes " an herbe which is sowed apart by itselfe, and is called by the inhabitants Upp6woc : In the "West Indies it hath diners names according to the seuerall places and countries "where it groweth and is vsed ; The Spaniardes generally call it Tobacco. The leaues thereof being dried and brought into powder, they vse to take the fume or smoke thereof by sucking it through pipes made of claie, into their stomacke and heade; from whence it pnrgeth superfluous nearne and. other grosse humors, openeth all the pores and passages of the
1 See Tab. 10, book iii., p. 17. London, 1705. 3 " Travels," etc., p. 296. London, 1778. 3 Page 16. Francoforti ad Mosnum. DC Cry, mo 1590.
396
ANTIQUITIES OF THE SOITTHEBN INDIANS.
body : by which raeanes the vse thereof not only preserueth the body from obstractios, but also if any be so that they haue not beene of too long continuance, in short time breaketh them: wherby their bodies are notably preserued in health, and know not many greeuous diseases, wherewithall woe in England are oftentimes afflicted. This TJpp6woe is of so precious estimation amongest tliem that they thiiike their gods are mamelously delighted therewith; Whereupon sometime they make hallowed fires, and cast some of the poiider therein for a sacrifice : being in a storme uppott the waters, to pacific their gods, they cast some vp into the airo and into the water ; so a weare for fish being newly set vp, they cast some therein, and into the aire; also after an escape of danger they cast some into the aire likewise ; but all done with strange gestures, stamping, sometime dauncing, clapping of hands, holding vp of hands, and staring vp into the heaueiis, vttering therewithal, and chattering strange words and noises,"
In the voyage of Sir Francis Drake, it is mentioned that some of the North American Indians " brought a little basket made of rushes, and filled with an herbe which they called Tobah ; " and Drake afterward adds : '' They came now the second time to us, bringing with. them as before had been done, feathers and bags of To7>a7i for presents, or rather indeed for sacrifices, upon this persuasion that we were gods." * Admitting the devotional, propitiatory, religious, political, medicinal, and social uses to which the pipe, in its various forms, was dedicated; conceding the divine origin claimed both for it and tobacco, and granting that around them clustered superstitions and ceremonies unique in
1 Quoted in Stevens' " Flint Chip?," pp. 318, 319. London, 1S70.
MANUFACTURE OF PIPES.
397
their character and powerful in tlieir influences, it is nevertheless true that among the North American (and particularly the Southern) Indians, smoking constituted, from the earliest times, a sensual enjoyment. Among their personal effects a pipe was frequently, if not always, reckoned; and the narcotic influences of to bacco were sought after with an avidity engendered only l>y confirmed habit.
The Choctaws raised tobacco to sucli au extent that they sometimes sold it to the traders. ^Vheii using it for smoking-, they mixed it with the leaves of two species of the oariaria, or of the liquidambar styraciftua. The pipe was in common use among them, and in some shape or other was the symbol of peace, friendship, and social conversation. The first civility offered by the JVI\iscogees to a stranger was a pipe, and this, -when accepted, was followed by " a dish of venison and homany." ' Speaking of the manufac ture of pipes by the Southern Indians, Mr. Adnir 2 affirms that " they make beautiful stone pipes; and the Cheerake the best of any of the Indians : for their mountainous country contains many different sorts and colours of soils proper for such uses. They easily form them with their tomohawks, and afterward finish them in any desired form with their knives ; the pipes being of a very soft quality till they are smoked with and used to fire, when they become quite hard. They are often a full span long, and the bowls are about half as large again as those of our English pipes. The fore part of each commonly runs out -with a sharp peak, two or three fingers broad, and a quarter of an inch thick.
398
ANTIQUITIES OF THE SOUTHERN INDIANS.
On both sides of the bowl," lengthwise, they cut sev eral pictui'es with a great deal of skill and labour; such as a buffalo and a panther oil the opposite sides of the bowl; a rabbit and a fox; and, very often, a man and a woman^wris naturalibns. Their sculpture cannot much be commended for its modesty. The savages wort so slow, that one of their artists is two months at a pipe Tvith his knife, before he finishes it; indeed, as before observed, they are great enemies to profuse sweating, and are never in a hurry about a good thing. The steins are commonly made of soft wood about two feet long, and an inch thick, cut into four squares, each scooped till they join very near the hollow of the s'tem; the beaus always hollow the scruares, except a little at each corner, to hold them together, to which they fasten a parcel of bell-buttons, different sorts of fine feathers, and several small bat tered pieces of copper kettles hammered round, deerskiix thongs, and. a red painted scalp : this is a boast ing, valuable and superlative ornament. According to their standard, such a pipe constitutes the possessor a grand beau. They so accurately carve or paint hiero glyphic characters on the stern, that all the war actions and the tribe of the owner, "with a great many circum stances of things, are fully delineated."
When Lieutenant Timberlake, in 1761, was pre sented to the Cherokees he was complimented with many professions of friendship and a string of beads. The pipe-dance was celebrated in his honor. The bowl of the pipe used on this occasion " was of red stone, curiously cut with a knife."
He saw other pipes made of black stone, and some inanxifactured from " the same earth they make their pots with, but beautifully diversified. The stem is
MAH0FACTUKE Of PIPES.
399
about three feet long, finely adorned with porcupine quil]s, dyed, feathers, deer's hair, and such like gaudy trifles." Having smoked the peace-calumet, he adds, " I was almost suffocated with the pipes presented me on every hand, which I dared not to decline. They might amount to about 170 or 180; which made me so sick that I could not stir for several hours." '
L<awson informs us that among the Carolina In dians the women were addicted to smoking. " They have pipes," says he, " whose heads are cut out of stone, and will hold an ounce of tobacco, and some much less." The same author perpetuates the fact that by these Southern Indians tobacco-pipes were mamifactured of clay with the express object of trans porting them into distant regions and there exchanging them with other Indians for raw skins, etc." Shortly after their primal intercourse with the -whites, the redmen, in many localities, appear to have adopted the shape of the European pipe, as being more convenient than, that formerly in vogue with them. As a matter of history it may be stated that the common clay pipe of commerce was, immediately upon its introduction, eagerly sought after by the Indians ; and thus it came to pass that those who were visited by the traders, or who enjoyed facile communication with the colonists, speedily abandoned the general manufacture of pipes, retaining, however, their calumets, and perpetuating the different ceremonies, uses, and traditions with which they -were so intimately associated.
In the olden time the Indian evidently laid great store by his pipe. For its construction the choicest material was often selected. This he collected not in-
1 " Memoirs," etc., pp. 38, 39. London, 1765. 2 "History of Carolina," pp. 55, 338. Raleigh reprint, 18GO.
400
ANTIQUITIES OF THE SOUTIIEEJT ISDIAJJS.
frequently at a great remove from his home, and in fashioning- and polishing the bowl days and months of labor and skill were consumed. Upon the stem, also, the ingenuity and taste of the owner were ex hausted. The presence of stone pipes iu mounds at a distance of sometimes several hundred miles from the locality whence the material of which they were manu factured could have been procured, fully attests the fact that a trade must have existed among the "abo rigines in these highly-esteemed articles. It has been more than hinted by at least one person whose state ment is entitled to every belief, that among the Cherokees, dwelling in the mountains, there, existed certain artists "whose professed occupation was the manufac ture of large stone pipes, which were by them trans ported to the coast and there bartered away in exchange for articles of use and ornament, foreign to, and highly esteemed among, the members of their own tribe. It will be readily observed that, in selecting materials for their stone pipes, the Indians chose such varieties as were best calculated to withstand the continued ac tion of heat. This "was undoubtedly the result of act ual experiment. In the absence of cutting and boring implements of metal, the construction of a pipe out of hard stone was a difficult and tedious undertaking. The constant and prolonged attrition required to re duce it to its desired proportions, the labor necessary for tracing the ornamental lines and hollowing out the bowl and the hole for the insertion of the stem simply with the aid of some rude nint implement, and the toil involved in imparting that degree of polish character istic of so many of the more elaborate pipes, were all known to the primitive pipe-maker. In order, there fore, to avoid, as far as possible, the chances of losing
CLASSIFICATION Ol' PIPES.
401
his pipes at an early day l>y their cracking under the influence of heat, he availed himself of the experience of his forefathers, and selected such varieties of stone as would best subserve his purpose, and at the same time most certainly perpetuate the results of his taste and industry.
The mound-pipes, described by Messrs. Squier and Davis,' exhibit a degree of art and skill unexcelled by any other specimens of ancient pipes fashioned by the North American Indians.
Passing from these general observations we turn to an examination of the antique pipes found within, the present geographical limits of Georgia. From the numbers taken from mounds, seen in refuse-piles and ploughed up in the fields, it may be confidently as serted that the Indians of this region were generally addicted to smoking. From the earliest historic pe riod the pipe was their almost invariable companion, and its intimate association w~ith the oldest monu ments proves that there was no epoch when its use was unknown to their ancestors. These pipes may appro priately be divided into three classes: First in inter est and in art is the IDOL-PIPE. This is rarely seen, and only in localities where, in the distant past, dwelt peo ples to all appearances more permanent in their seats and tribal organizations, more agricultural in their pur suits, more addicted to the construction of large tumuli, and superior in their degree of semi-civilization, to the nomads "who occupied the soil at the date of European colonization. Specimens of such pipes are as infre quent as stone images, and it is probable that they should both be referred in their origin to the handi-
' "Ancient Monuments of the Mississippi Valley," vol. i. ; " Sinitllsonian Con tributions to Knowledge," pp. 251-272. Washington, 1848.
402
ANTIQUITIES OF THE SOUTHERN TNDIAK3.
work and superstition of the primitive men who threw up those large mounds which tower along the banks of the Etowali and lift their imposing forms from out the level of several other valleys in Georgia. They are al ways associated, so far as our knowledge extends, -with the large pentagonal and quadrangular mounds, and with those older monuments--be they watch-towers, sepulchral tumuli, temples, consecrated spaces, enclosed areas, defensive works or play-grounds--of whose age and objects the later Indian tribes cherished not even a tradition. The best idol-pipes we have seen were ploughed up near the base of the pentagonal mound, within the enclosure formed by the moat and the Ktowah River, upon the plantation of Colonel Lewis Tumlin, near Cartersville, Georgia. A description of this interesting locality has already been given. Un fortunately, ail opportunity for presenting a proper account and of figuring those relics is now denied. During* the summer of 1850, the author enjoyed the pleasure of seeing three of these pipes at the residence of Colonel Tumlin. A_micl the devastations consequent upon the invasion of Georgia by the Federal armies, in 18G4, these, with other valuable relics, were either destroyed or carried away by the soldiers.
Writing from recollection, it may be stated that these particular idol-pipes were made, two of them of serpentine and the other of mica slate. They varied in height from three and a half to five inches, in breadth from two and a half to three inches, and in length from three to four inches. In each instance a human figure Avas represented in a sitting posture--knees drawn up --elbows resting upon the knees, and the extended hands presenting and clasping an urn-shaped, bowl. These bowls were about two inches in diameter, and,
IDOL--pir-Hs.
403
disguising the sex, rested upon the abdomen and lower part of the breast. The head, rising somewhat above the level of the top of the bowl, was thrown, back ward. The chin and forehead were both retreating ; eyes large and. upturned--ears prominent. The fore,head was low, broad and bald. The hair, collected from all sides, was" confined at the top of the head, and thence falling backward was gathered into a sort of knot below. The coxmtenances were decidedly idiotic, and yet the devotional idea was forcibly expressed in the attitude and general appearance of these rude idolpipes--incense offering to an unseen yet acknowledged Deity. To these exhibitions of his skill tlie primitive sculptor had imparted a considerable degree of pol ish. The perforation for the stem passed below the shoulders through the back and belly into the bottom of the bowl. At its inception it -was three-qnarters of an inch in diameter, gradually lessening as it deep ened, until, at the point where it entered the bowl, it was scarcely a quarter of an inch, in -width. These pipes were obviously "very old; arid in all likelihood antedated, by an indefinite period of time, the occu pancy of this valley by the Cherokees. So far as re corded observation extends, nothing like them was noted in the use or possession of the modern Indians. There are at least plausible grounds for believing that the ancient peoples who piled up these august tumuli along the banks of the Etowah, and departing left be hind them, enduring monuments of their combined labor for a wonder and an enigma to later tribes, may have borrowed gome of their ideas of sun-worship, idolatry, agriculture, and of art directly or indirectly from the Southern cradle of American, civilization.
In the second class we include CALUMETS and large
404
ANTIQUITIES OF TIIK SOUTHERN INDIANS.
pipes whose size suggests the impression that they did not generally accompany the owner, but were carried only on special occasions, and used when prescribed ceremonies of a political, religious, medicinal, or war like character were to be observed. Varying in form and weight, such pipes are found both in the fields and in mounds. As a general rule, the more remarkable of them may be regarded as the public property of the tribe; still, their presence in conical earth-mounds con taining but a single skeleton, would seem to indicate that some of them were the private property of noted personages--perhaps chiefs and medicine-men. It is scarcely probable that the public peace-pipe would have been liable to inhumation. Among the most curious of this class are the bird-shaped pipes of which Figs. 1, 2, and 3, Plate XXIII., may be regarded as in teresting specimens. The first (Fig, 1) made of ser pentine, is seven inches and a half from the tip of the beak to the end of the tail, three inches in height to the top of the bowl, and. two inches and two-tenths in width just in rear of the bowl. The bowl is an inch and a half in diameter at the top, circular in shape, and an inch and five-eighths in depth. The walls of the bowl are three-eighths of an inch in thick ness. The aperture for the stem, commencing under the tail, passes longitudinally through the body of the pipe until it intersects the bowl at its bottom. At its inception this aperture is an inch in diameter, grad ually lessening as it deepens, until at the point where it communicates with the bowl it is only a quarter of an inch in width. The weight of this pipe is nearly two pounds.
The length of the second (Fig. 2) does not vary a quarter of an inch from that of the first. Its -weight
f i
L I ', f
'
I [
( | [' ' ( \ 'i-
CALUMETS ASD BIRD--SHAPED PIPES.
405
exceeds, however, by rather more than a quarter of a pound, and its height is three inches and three-quar ters. The shape of the bowl is elliptical--its walls being half an inch thick. In depth this bowl meas* ures two inches and a quarter. Its diameters at the top, reckoned respectively in the direction of the ma jor and minor axes of the ellipse, are an inch and a half and an. inch and three-tenths. The perforation for the stem is also elliptical, its greatest and least di ameters being six-eighths and five-eighths of an inch. This aperture is half an inch in diameter where it enters the bowl. This pipe is made of serpentine, and, like the former, has been carefully polished.
The third pipe (Fig. 3) is of oolite, of a creamcolor, and weighs two pounds and a half avoirdupois. It is six inches and a half in length, and about four inches high. The walls of the bowl are half an. inch in thickness, and the bowl--circular in form--possesses a diameter of an inch and three-eighths, and a depth of two inches and a half. In rear of the bowl this pipe is nearly three inches in thickness. The aperture for the stem is rather more than three-quarters of an inch in diameter at its inception below the tail, and is diminished to half an inch at the point where it enters the bowl. While resting upon the flattened beak, breast, and clumsily-represented legs, this seems only a, bird-pipe. If, however, we change the position, placing it upon the feet and tail, and turning the bowl away from us, this pipe at once assumes an entirely different aspect, apparently foreign to its ordinary uses, arid seems to assert its right to be classed among the idol-pipes. This modified view is presented in Fig. 4, Plate XXIII. This pipe was found in. the Chattahoochee Valley, several miles below the city of Colum-
406
ANTIQUITIES OF THE SOUTHMEIf IUDIAK"S.
bus. The other two pipes were taken from mounds, one in ISibb County, and the other in Greeiie. The cluck seemed to be the bird which most frequently enlisted the imitative powers of the Indians of this region. At other times the bear and the cougar, and the " human face divine," engaged the skill of the primi tive artist.
Of the ordinary forms of calumets, Figs. 5, 0, 7, 8, and 9, Plate XXIII., may be regarded, as typical. Fig. 5 consists of a stone composed of mica and dark brown felspar. This pipe is five inches high, nearly four inches long, and an inch and a quarter wide at the bottom, which is entirely flat, so that the pipe readily remains in an upright position. The bowl is circular, its diameter at the top being an inch and a quarter, and its depth rather more than three inches and a half. The walls of the bowl are three-quarters of an inch thick. The aperture for the stem is also circu lar, and three-quarters of an inch in diameter at the opening.
Number 6, composed of gneiss, is five inches in height, and five inches in length. The bowl is square, or very nearly so, the length of each side being about an inch and a quarter. Six-tenths of an inch will ex press the average thickness of the walls at the top. The bowl is rather more than three inches and three quarters in depth. This pipe also readily maintains an upright position, being flat at the bottom, which is an inch and three-quarters wide. The opening for the stem is circular, and, at its beginning, is an inch and a quarter in diameter. Fig. 7 represents a very fine calumet of steatite, four inches and three-quarters in height, foui* inches and a half in length, and about two inches wide. The circular bowl is two inches in
CALUMETS.
407
diameter, and the walls are not less than half an. inch
in thickness. The bottoms of all tliese pipes are flat,
and the bowls, with the exception of Fig. 9 (of soap-
stone), are at right angles with the stem.
f
Many of the pipes of this class arc made of mica
slate and soap-stone. The latter material being easily
worked and generally accessible, seems to have been
held in especial esteem. Varioris are the devices and
I
ornamentations traced upon the sides and faces of the
I
soapstoue pipes. Fig. 8, Plate XXIII., furnishes ail
i
example in point. Upon its bottom the paw of a bear
['
is traced. In front are square, circular, elliptical, and parallelogrammic figures, and the upper portion and sides are ornamented with various incised lines. These
i\
calumets were taken from mounds and ploughed up in
the fields.
Thus far the writer has failed to discover a single
instance of the use, among the Georgia Indians, in an
cient times, of the genuine red pipe stone or Gatlinite.
In the case of tlie softer stones there are indications
that the bowls and holes for the stem were made by
boring with a triangular-shaped implement, probably
of flint. Upon some of the inner surfaces of these
openings are annular abrasions, gradually decreasing
in diameter as the end of the aperture is neared. Cat-
lin 1 tells us that the Indians of the West shaped the
bowls of their pipes from a solid stone, not quite as
hard as marble, with nothing but a knife. " The
stone," he continues, " which is of a cherry red, admits
of a beautiful polish, and the Indian makes the hole in
the bowl of the pipe by drilling into it a hard stick,
shaped to the desired size, with a quantity of sharp
408
ANTIQUITIES OF THE SOUTHERN" INDIANS.
sand and water kept constantly in the hole, subjecting him, therefore, to a very great labour and the necessity of much patience."
In his account of the manners and customs of some of the "Western Indians, Mr. Hunter l writes : " The men occasionally amuse themselves with, making bowls and pipes of clay, for their individual use, which are burned as before described. They also make bowls and pipes of a kind of indurated bole, and of compact sand and limestone which are excavated and reduced to form by means of friction with harder substances and the intervention of sand and "water. They generally ornament them with some figure characteristic of the owner's name; as, for instance, with that of a buffalo, elk, bear, tortoise, serpent, ete., according to the circum stance or caprice that has given rise to its assumption. Ill the same way they manufacture their large stone mortars for reducing corn into fine meal." In the ab sence of all metallic implements, it is probable that the Southern Indians gave outward shape to their harder pipes mainly by means of attrition. Sharp sand and water may have materially assisted them in drilling the holes. The drill, in all likelihood, consisted either of a piece of hard wood or of cane. As the cavity of the bowl narrowed, smaller drills were employed until the bottom was reached. Professor liau, in his excellent article upon drilling in stone without metal, advances the opinion that a piece of cane will form " a regular hollow cylinder sufficiently strong to serve as a drill." * In this belief I fully concur, and am firmly
USE OF SOLID AXD HOLLOW DRILLS.
409
persuaded that the hard, cane (^rundinaria macrosperma, Michaux), furnished tlie Southern Indians, and that abundantly, with hollow drills, which, with the aid of sharp sand and water, and a liberal expenditure of time and labor, would have compassed the perfora tions and hollows we observe in these pipes. Dr. Davis informed Professor Ran that a stone pipe, with an unfinished hollow partly filled with vegetable mat ter, was sent from Mississippi to the late ]Dr. Samuel Gr. Mortoii, of Philadelphia. When subjected to a mi croscopical examination, this vegetable substance ex hibited the fibrous structure of cane, and thus appeared to be the remnant of a drill broken off in the bore. 1 Some of the bowls of the soapstoue pipes were evi dently hollowed or dug; out with the aid of a sharppointed nint implement. Instances occur where the workman, neglecting to smooth or polish the inner surface, has left the marks of his rude incisive instru ment. In many of these pipes the apertures for the stems appear unnecessarily large ; and yet, the size of those openings, in connection with their flat bottoms, furnishes an argument in support of the principal use to which we suppose them to have been dedicated. In the deliberations of the council-lodge, or upon pub lic occasions, it was important that the decorated stems should be long enough to be conveniently passed from one to the other, as the chief men sat around, without lifting the pipe from the ground upon which it rested. To accomplish this object, and also to afford ample op portunity for that labored ornamentation which was the pride and boast of the red-men, the stem must have been large and long. It will be remembered that the office of pipe-stem-carrier was, among many of the
3 Animal Report of the Board of Regents of the Simtlisonian Institution for 1868, p. 399.
410
ANTIQUITIES OF THE SOUTHERN INDIANS.
tribes, invested with no little dignity. Swamp-canes furnished ample and convenient material, among the Southern Indians, for the facile manufacture of stems of any desired size or length. Joints of graduallydiminisliing diameters could be readily adjusted the one into the other. All necessity for perforation was avoided, and thus a tapering stem, light and strong, conld Ite easily constructed whose larger end would fit the aperture in the pipe, while its smaller extrem ity w-onld admirably answer the convenience of the smoker.
The third class includes the OKDINAKY PIPES in common use among the natives for smoking tobacco and other leaves, weeds, and barks, "whose narcotic properties were well known to them in their primitive state. The luxury of smoking from the earliest times was recognized by nearly all the American tribes, "There is no custom," says CatKn,' "more uniformly in constant use amongst the poor Indians than, that of smoking, nor any other nioi^e highly valued. His pipe is his constant companion through life--his messenger of peace; he pledges his friends through its stem and its bowl, and when its care-drowning fnmes cease to flow, it takes a place with him in his solitary grave, with his tomahawk and w^ar-club, companions to his long-fancied, ' mild, and beautiful hunting-grounds.' "
These common pipes were made both of stone and clay, generally of the latter material. They are usually of a size capable of being easily transported, and are not much heavier than the ordinary pipe of the present day. Some are no bigger ill the bowl than a thimble. Of the stone pipes, Figs. 2 and 6, Plate XXIV., may be
7>(-a te^KIS.
COMMON" nPES OF STO^E ANT} CLAY.
41 1
regarded as fine specimens, perhaps rather more deli
cate than those in common use. The pipe represent
ed by Fig. 2 was found near the large mound on the
plantation of Mr. J. H. Nichols, in Naeoochee Valley.
Made of a hard Mack stone, it has been formed with
much, regularity and delicacy. The walls of the bowl
are very thin, scarcely thicker than the sides of a sad
dler's thimble.
The composition of the clay pipes is precisely the
same as that iised by the Indians of this region in the
manufacture of their pottery, red arid blue clay mixed
with powdered shells or fine gravel.
Figs. 4 and 7, Plate XXIV., represent two pipes
of this description taken from, earth-mounds on the
Ocmulgee Kh-er, not far from Macoii; while those
delineated in Figs. 3 and 5, Plate XXIV., were found in
shell-mounds on the Colonel's Island. Fragments of
pipes of this composition are not infrequent, and attest
their general use among the ancient inhabitants of this
region. Perfect specimens are rarely to be obtained.
The custom of burning the dead was, at some time
or other, maintained to a considerable extent on the
Southern Atlantic coast. A.S a direct consequence, in
all tumuli where cremation occurred, only fragments of
pipes may now be found. From one small mound of
this character the writer obtained parts of five clay
pipes which had been broken iu the funeral-fires.
Hearne describes a custom among the Chippewas, after
the shedding of blood, of throwing all their ornaments
and pipes into a common fire; and "Winslow narrates
of the IV^anohiggausets that they had a house, ordinarily
t
frequented by priests, whither at certain times resorted
all the people and offered their riches to their gods.
These contributions were cast by the priests into a
412
AXTIQUITIK3 O_F THE SOUTHERN INDIANS.
great fire made in tlie middle of that house. 1 Upon tlie sacrificial altars of the mounds of tlio West, Messrs. Squier and Davis found many beautiful pipes cracked and broken by fire. It may be that these fragmentary clay-pipes from the Greorgia mounds, if rightly under stood, testify tlie sincere affection cherished ~by the living for the dead when, having concluded the last funeral rites, they committed to the same fires which consumed the bones of the departed, these symbols of peace, of comfort, and of friendship.
Figure 1, Plate XXIV., is a correct delineation of a clay pipe found in a grave-mound in Tennessee.
The modern Cherokees excelled ill tlie manufacture of bird and. animal-shaped pipes--many of them large and elaborate. Tlie nude human figure in kneeling, bending or sitting posture, frequently formed the sub ject of imitation; and we have seen several pipes of this description which, in the language of ^Vdair, could not " much be commended for their modesty."
' Sec Wilson's " Prehistoric 3Isn," second edition, p. 323. London, 18u5.
CHAPTER XIX.
Idol-Worship among the Southern Indiana.--Stone imd Tevrn-CoUa Image*.
TJIE history of idol-worship--from its most de graded expression in the Fetichism of Congo, through all its modified forms up to its most elaborate develop ment iii the states of ancient Greece and Home--is both curious and interesting. The stocks and stones and the senseless images of the unlearned and the base have perished, and are passing into oblivion wherever the shadows of superstition, beneath which they had their being, are dispelled by the light of a superior civilization. Even the bulls and beetles of enigmatic Egypt--overrun with gods--incurred the sneers of Juvenal. Although not a single worshipper be found among living men, the divinities of Olympus, the Muses, the Graces, the Lares and Penates, the Fates, the Fu ries, and the Satyrs of the classic age, and the sublime art which enthroned them, on earthly pedestals, still live in the domains of literature and taste. They are as immortal as the poetry, the imagination, and the traditions, whose offspring they were. Beautifully has Coleridge testified to the permanency of this art idolatry, and to the iufluence which its memories still exert over the minds of succeeding generations :
414
A^i'TIQUITIKS O'F TETB SOtTTJIKBW
" Tlio iniclligible fomxs of ancient poets, The fair humanities of old religion, The Power, the Ueautj, arid tlio Majesty That liad their haunts in dale or piny mountain, Or forest, l>v slow stream, or pebbly spring, Or chasms and watery depths; all these have vanished; They live no longer in the faith of reason; X>nt still the heart clotli need a language ; still Doth the old instinct bring back the old names ; Spirits or g'ocls that used to share this earth "With man as v.ith their friend; and at this day 'Tis .Jupiter who brings whate'er is great, And Venus who brings every thing that's fair."
But it is not permitted, us now to linger among; these deifications of the unseen powers of Nature-- these wonderful embodiments of the ideal and the "beautiful. To humbler and more obscure investiga tion ft do OUT present inquiries lead.
Mr. Tylor * is of the opinion that idols belong- to a period of transition and of growth. In support of this idea, he instances the fact that among races lowest in the scale of civilization--such as the _Fuegians and many of the Indian tribes of TSTorth America--we see and hear little or nothing of idols, while in Mexico and Pern the entire apparatus of temples, idols, priests, and sacrifices, obtains in a complex and elaborate form. A belief in the existence of a Supreme Being is wellnigli universal among' men, and the absence of all religions superstitions and. of a conception of the immortality of the soul, is the emphatic sign of the most absolute degradation. The presence of idols among barbarians may be therefore regarded as denoting not only the entity of a religious idea, but also the cooperation, of
' " Researches into the Early History of Mankind,'' etc., second edition, p. 312. London, 1870.
HXXL--"WORSHIP.
41 O
something like art and imagination to impart definite shape and personality to vague conceptions of superior beings.
Sympathizing with the views of jVTr. Tylor, Sir John X/ubbock l writes: " The worship of idols charac terizes a somewhat higher stage of human develop ment. "We find, no traces of it among the lowest races of men ; and Lafitau says truly: ' On pent dire en general que le grand nombre des peuples souv ages n"a point d'idoles.' The error of regarding Idolatry as the general religion of low races, has 110 doubt mainly arisen from confusing- the Idol and the Fetich. Fetichism, however, is an attack on the Deity ; Idolatry is an aet of submission to him j rude, no doubt, but yet Immble. Hence, Fetichisni and Idolatry are not only different, but opposite, so that the one could not be developed directly out of the other. AVe must, there fore, expect to find between them, as indeed we do, a stage otfc religion without either the one or the other."
However true it may be that idol-worship indicates a development of the religious idea as contrasted with its non-exi.stence among peoples who give evidence either of no religious emotions whatever, or of super stitions so degraded that they do not rise above Fetichism, certain it is that a devotion which, ignoring the intervention of idols, recognizes the existence of a Supreme Being, a Great Spirit, or of two controlling divinities--the one of good, and the other of evil--is still more elevated and expansive in its character. So also is tliafc system, of worship which, deifies the sun and moon, and cherishes fire as an object of adoration because of its supposed direct emanation from a divine
n of Civilization and the Primitive Condition of Man," p. 225.
416
ASTIQriTtEi OF Til]! SOt'TtlERS TKDIANS.
source. The religion of the Southern Indians was, in some respects, not unlike that attributed by Tacitus to the ancient Germans : " Cffiterum nee cohibere parieti' bus deos iieque iu uliam humani oris ypeciem assimulare, ex magmtudine caslestium arbitrautur. Lucos ae nemoraconsecrant, deorunupie nomhribus appellant secretum illud quod sola revcreiitia vident." '
Both the sun and moon -were, among many tribes, regarded with absolute 'veneration ; and certain seasons were set apart for special religious observances. Speak ing generally, it may be stated that they recognized one great ami good Spirit as the creator of all--the author of life, and light, and lieat--tlio dispenser of rain, the provider of game, and the source of all devel opment in plant and animal. Him they sought to pro pitiate on all important occasions, whether of war, the chase, or of husbandry. Subordinate to this great, first cause, they reckoned other spirits, g-ood and evil, and with them their 'priests, conjurers, and medicine men were commissioned to treat. ^\rith the malign influences of the evil one--whether exerted iu the form of disease, or faint-heartedness, or blight upon the zea --they were ever contending. Extravagant as were their traditions and stiperstitiona with regard to their national or tribal origins, there vras always incorporated, some memory which perpetuated the primal presence and power of this Great Spirit. His intervention was admitted in the first strong wind, great fire, or dense smoke, or in the opening of-some vast cave from which his children issued forth to possess the green earth ho had made. Despite their curious and degraded re ligious notions, there can be no doubt but that many
1 C. Oornelii Taciti Opera omnui, ,1:1 fidjm eJitionia Oi-clliana:,'' toin. ii., 1>. 235. Oxonii, Igol.
OKORGTA TIMBKS JTOT IDOL--WORSHIPPERS.
417
of the tribes, realizing the presence of a soul or spirit in tlie breast of man, and appreciating tlie operation of natural laws, attained unto a conception not only of the immortality of that spirit, but also of the fact that in a future state good or evil fortune would betide the translated according to his conduct in this world.
That the Georgia tribes were not idol-worshippers --in the ordinary acceptation of that term--and did not fashion or reverence images at the period of our earliest acquaintance with them, may be confidently
affirmed. Speaking of the Indians who resided in the vicinity
of Savannah when General Oglethorpe established the colony of Georgia beneath the pines which then crowned Yamacraw Bluff, the lieverend Mr. Bolzius ' states: " They have some Religion, believing a Su preme Being, which they call Sotolycate (literally translated, Jfe w7to sitteih, A-l>ove,), who is in all Places ; though they wonld not teach us the AVord by which they express the Name of GOD in their Language. They believe that from the Supreme Being comes every Thing, especially Wisdom. They use no Cere monies, nor ontward religious lixercises, except at a Solemn Festival, held once a Year. They worship no Idols ; however they sing some songs about the ancient
Heroes." Equally emphatic is the testimony of Mr. Bartram: *
"These Indians are by no means idolaters, unless their puffing the tobacco smoke towards the sun, and rejoic ing at the appearance of the new moon, may be so
1 An "Extract of the Journals of Mr. Commissary Von Rock anil of the Rev erend Mi'. P-olxius, p. :j(5. London, 1734.
a " Travels through North and South Carolina, Georgia, East and West Florida," etc., pp. 495, 490. London, 1*792.
27
418
ANTIQUITIES OF TILE SOIJTTTEISN INVDIAKS.
termed. So far from idolatry are they, that they have 110 images amongst them, nor any religious rite or cere mony that I could perceive ; but adore the Great Spirit, tlie giver and taker away of the breath of life, with the most profound and respectful homage. They be lieve in a future state where the spirit exists, which they call the world of spirits, where they enjoy differ ent degrees of tranquillity or comfort, agreeably to their life spent here; a person who ill his life has been an industrious hunter, provided well for his family, an intrepid and active warrior, just, upright, and who has done all the good he could, will, they say, in the world of spirits, live in a warm-, pleasant country, where are expansive, green, flowery savannas and high forests, watered with rivers of pure waters, replenished with deer and every species of game; a serene, unclouded, and peaceful sky; in short, where there is fullness of pleasure, uninterrupted."
Mr. Adair * is no less positive in his recorded obser vations on tliis point: " But these Indian Americans pay their religious devoir to Loak-Ishto Jwollo--Aba, 'the great, beneficent, supreme, holy spirit of fire,' who resides (as they tliink^) above the clouds, and on earth also, with unpolluted people. He is with them the sole author of warmth, light, and of all animal and vegetable life. They do not pay the least perceivable adoration to any images, or to dead persons ; neither to tlie celestial luminaries, nor evil spirits, nor any created being whatsoever. They are utter strangers to all the gestures practised-by the pagans in their re ligious rites. They kiss no idols; nor, if they "were placed out of their reach, would they kiss their hands in token of reverence and a willing obedience. . . .
1 ' History of the American Indians," etc., pp. 19, 22. London, 1775.
IMAGES.
419
They P&y so religions worship to stocks or stones after the manner of the old Eastern pagans ; neither do they worship ally kind of images whatsoever. . . .
" I never heard that any of our l^orth American In dians had images of any kind. There is a carved human statue of wood, to which, however, they pay no religious homage. It belongs to the head war-town of the Upper Muskohge country, and seems to have been originally designed to perpetuate the memory of some distinguished hero who deserved, well of his country; for, when tlieir c'ussee'ii.a, or bitter black drink, is about to be drank in the synedrioii, they fre quently, on common occasions, -will bring it there, and honour it with the first conch-shell-full, by the hand of the chief religious attendant : and then they return it to its former place. It is observable that the same beloved waiter, or holy attendant, and his coadjutant, equally observe the same ceremony to any person of reputed merit in that quadrangular place. When I past that way, circumstances did not allow me to view this singular figure ; but I am assured by several of the traders, who have frequently seen it, that the carving is modest, and very neatly finished, not un worthy of a modern civilized artist." The same author assures us that he has never seen the "worship of the Priapxis indulged in by the natives with whom he was accruainted.
Referring to the Cherokees, Lieutenant Timberlake ' says: " As to religion, every one is at liberty to think for himself; whence flows a diversity of opinions amongst those that do think, but the major part do not give themselves that trouble. They generally concur, however, in the belief of one superior Being
1 " Memoirs," etc., pp. G3-G5. London, 1VG5.
420
ANTIQUITIES OF TIIK SOUTHEK^T IKDIAKS.
wTio made tliem and governs all tilings, and. are there fore never discontent at any misfortune, because tliey say tlie Mini above would have it so. They believe in a reward and punishment, as may be evinced by their answer to Mr. Martin, who, having preached Scripture till both his audience and he wore heartily tired, was told at last, that they knew very well that if they were good they should go up ; if bad, clown ; that lie cotilcl tell no more ; that he had long plagued them with what they no ways understood, and that they desired him to depart the country. . . . They have few religions ceremonies or stated times of gen eral worship : the green-corn dance seems to be the principal, -which is, as I have been told, performed in a very solemn manner in a larg'e square before the townhouse door : the motion, here is very slow, and the song in which they offer thanks to Grotl for the corn he has sent them, far from nnpleasing-."
"When questioned, as to the origin of the new "fire, and of the 15oos-ke-tau? ICfau Haajo, 1 the great- Medal Chief of Took-aiz-bat-che, responded that he had been. taught from his infancy to believe that there is an 3vsau-ge-tuh E-mis-see (^Master of Breath) who gave these customs to the Indians as necessary and suited to them; and that an observance of them entitled the red-men to his care and protection both in war and seasons of difficulty.
When asked whether the Creeks believed in a future existence, he replied.: " The old notion among1 us is that 'when we die the spirit Q>o-yau.-fi.c-chau-) goes the way the sun goes, to the west, and there joins its family and friends who went before it." To the in-
HEGARDIXa A SUPK'ElrK BMSG.
421
quiry, " Do the red people believe in a future state of rewards and punishments ? " lie answered : " Wo have an opinion that those who behaved well, are taken uiider the care of E-sau-ge-tuh E-mis-see and assisted ; and that those who have behaved ill, are left there to shift for themselves ; and, that there is 110 other punish ment."
During a conversation which occurred between. Tomo-chi-ehi and G-eneral Oglethorpe about prayer, the aged jMico of the Yamacraws said that the Indians never prayed to God but left it to Him to do what He thought best for them : " That the asking for any par ticular blessing looked to him. like directing G-od ; and if so, that it must be a very wicked thing. That for his part he thought everything that happened iu the world was as it should be; that God of Himself would do for every one what was consistent with the good of the whole ; and that our duty to Him was' to be con tent with whatever happened in general, and thankful for all the good that happened in particular." *
In. his " Philosophico-historico-hydrogeography of South Carolina, Georgia, and East Florida," De Brahm* savs that the .Indians of this region entertained a no tion of immortality and of a future state wherein, they expected to enjoy large hunting-grounds well stocked with deer, and also an apprehension of spirits. " The Indians have also a very scant knowledge of a Drviiie Being, which knowledge, or rather notion, extends no farther than that they believe he is good : the Cherakees call him (Hianequo), the great man, whom the
1 Spenee's "Anecdotes," p. 318. London, 1820. Jones' " Historical Sketch of Tomo-chi-chi," p. 103. Albany, 1868.
5 " Documents connected with the History of South Carolina, edited by Plowden Charles, Jennett Weston, and printed for private distribution only," pp. 221, 222. London, 1836.
422
AUTIQUITIKS Or THE SOUTH ET!S IKDIANS.
Catabaws call (Rivet), overseer; but they pay no man ner of adoration to him, nor anything existing; nor have they any ceremony at all than to extinguish all their fires once a year in July, at the time when the Indian corn (maize) is in its milk, vrhich they squeeze out l>y beating and. straining; than boile that milk by a fire new caught from electrif'ation, "which they per form with two green sticks rubd with great velocity a cross each other until they are lighted ; when this milk is boiled to a consistency, they let it cool, than form it into little cakes which they fry in bear's fat, and are (whilst warm) a delicious eating; with them they keep feasting three days. To this season they postpone all elections, promotions, and their king's coronations."
" Their Religion,1' upon the authority of Mr. Ash,' , " chiefly consists in the Adoration of the Sun, and Moon : At the Appearance of the New J\foon I have observed them with open extended Arms, then folded, with inclined Uodic*, to make their Adorations with much Ardency and Passion."
If we may credit the narrative of Jonathan Dickeiison," a sort of Sabiauism existed among some of the Florida tribes, and Pitchlyiin once remarked : " from all I have seen and. can understand of the Indians who once inhabited the portions of country covered by the Southern States of the Union, they appear to have been originally worshippers of the sun. The Chahta -when he hag greatly misbehaved, utters these ejacula tions : "When the sun forsakes a man he will do things he never thought to do ! The sun is turned against me, therefore have I come to this." G The Kdalgo of
1 " Carolina," etc., by T. A., Gent., p. 30. London, 1682. - lt (loot's Protecting Providence," etc., third edition, p. 13, cl alitcr. Phila delphia, 1720. 3 " Relation of Alvar Kufioz Cubeca de Vaca," translated by Buckingham Smith, p. 171, note 3. New York, 1871.
woowEjr STATUES;
423
Elvas asserts tliat tlie Indians of Florida worshipped the devil, and made sacrifices of the blood and bodies of their people whenever his Satanic majesty suggested that lie was athirst. It was to escape such oblation, that Juan Ortiz, warned by the Indian girl, fled by night to JMococo. 1 Cabeea de Vaca mentions " gourds bored with holes and having pebbles in them," which were used by the Indians in their dances, and were supposed to possess special virtue because they were heaven-descended."" Elsexvhere in the Spanish iiarratives do we read of wooden images of birds ; but, so far as we now remember, no account is given of a single idol as an object of adoration among the aborigines. At Talomeco, De Soto found a large temple or mauso leum, at whose entrance were gigantic statues of wood, carved with considerable skill, the largest of which was twelve feet high. They "were armed "with various weapons, and " stood in threatening attitudes, with ferocious looks." The interior of the temple was decorated with statues of various shapes and sizes. There was also a great profusion of conchs and di/fereiit kinds of sea and river shells. It does not appear, however, that these images were objects of religious veneration or positive worship. Like the " carved human statue of wood " in the head war-town of the upper Muskohge country, described by A_dair, they seem rather to have been the effigies of heroes, the embodiments of brave memories, the symbols of tribal pomp and power.
Lawsoii tells us that at the corn-dances among the Carolina Indians--the one when the harvest is ended, to return thanks to the good spirit for the fruits of the
421
AXTIQTTITIES OF THE SOUTHERN IKDIAKS.
earth--the other in the spring to invoke blessings upon, the seed to be sown--the old men, in order to encourage the young men to labor stoutly in planting their maize, set up a sort of idol in the field, attired in the customary habit of an Indian, with strings of wampum about its neck. This image--guarded by the king and old men, who pay profound respect to it--the young ones are not permitted to approach. By the old men the young men are told that this image " is some famous Indian warrior that died a great while flgo, and now is come amongst them to see if they will work -well, which, if they do, he will go to the good spirit and speak to him to send them plenty of corn," and make them " expert hunters and mighty war riors." ' Commenting upon the absurdities of their superstitions, our author asserts it to be impossible to give a true description of their religion : "I * have known them," says he, " amongst their idols and dead kings in their Quiogozon for several days, where I could never get admittance to see what they were do ing, though I was at great friendship with the king and great men: but all my persuasions availed me nothing, neither were any, but the king, with the con jurer, and some few old men, in that house ; as for the young men and chiefest numbers of the Indians, they were kept as ignorant of what the elders were doing, as myself. They all believe .... that there are two spirits ; the one good, the other bad. The good one they reckon to be the author and maker of every thing, and say that it is he that gives them the fruits of the earth, and has taught them to hunt, fish, and be "wise enough to overpower the boasts of the wilderness,
1 "History of Carolina.," pp. 2S5, 23. Reprint. Raleigh, 1830. 2 Idem, p. 3-t2.
AJTCIENT GODS OF THE VIRGINIA ISDIAIfS.
42 O
aiid all other creatures tliat they may be assistant and beneficial to man." They declare also that the bad spirit lives apart from the good, and torments men with siclvness, disappointments, losses, hunger, and all the mis fortunes incident to human life. In the immortality of man they believe, and. have a notion of certain re wards and punishments in another -world.
Beverly' furnishes an account of a surreptitious visit which he paid to a Quioccosan, or house of reli gious worship, used Toy the Virginia Indians. Ill a mat he there found what he took to be a disjointed idol--a rude affair, scarcely justifying the elaborate representation offered in the accompanying; plate. He also assures us that these peoples had altars and places of sacrifice. To the evil spirit burnt-offerings were made, and it is more than probable that on some occa sions young children were immolated.
- Treating of the religious belief and worship of the Virginia Indians, the author of the " Admiranda Narratio " " says : " jMultos Deos crcdunt, quos JVToNTOAc appellant, d.iuersorum tamen genevum & ordinum : unuin solnm primarium & Magnum I3eum o[ui fuerit ab seterno. Is (ipsis afferentibus) miindum conditurus, initio creauit alios decs primarii ordinis, ut essent tamo^nam media tfe instrumenta, ipsi subseruientia cuin ad creationeni, turn ad gubernatioriem : deinde Solern, Lunam & Stellas tamquam Semi-Deos & instrumenta alterius ordinis prcecipui. Dicunt aquas prbuum om nium esse factas, ex quibtis Dii omnes creaturas visibiles & in\dsibiles condiderunt. . . .
" Omnes Deos litirnanam naturam habere putant,
426
ANTIQUITIES OF THE SOUTHER^T IXDIAK3.
ea de causa imaginibus humanae formas illos expriintmt, eosque KEWASOWO.K appellant, unicus KEWAS est clictus. Illis proprias cedes sine teinpla dicant, quo3 Machiconmck nominat, in quib, sint precationes, cantus <fe.per mnltos dies oblationes ad ipsoru Deos. In quibusdam templis iios TinicTiin ICe^vas obseruauimus, in aliis biuos, aliqnando tres. Vulgas etiam pro Diis liabet.
" A.ninifB imniortalitatem. ol credunt, earn statim atque a eorpore soluta est transferri secudura opera qnso fecit, vel ad Deorum aedes ad perpetuam felicitatem trueudam, vel ad ingontem fossam seu Bcrobem (quam in extremis muiidi finibus procul ab ipsis versus occidentem sitis esse ceiisent) ad perpetuinu ignem: euni locum ipsi POPOGUSSO appellant."
In plate xxi., De Bry presents us with a sketch of the idol Kiwasa seated in its temple. The illus tration is accompanied by the following explanatory remarks : " Idoruni habent huius regionis incola3, KIWASA appellaturn, e ligneo trunco elaboration, quatnor pedes alfctim, ^ cuius caput Floridae incolarum capita refert: fades cameo colore depiota est, pectus aTbo, rcliqunrn corpus nigro, crura etiam picfcura alba variegata: e collo torques pendent Hphseriilis albis coristantes, qni"bus intermixtiB sunt, aliaj teretes ex tere, magis ab illis oastimato, quam aurum vel argentum. Est illud idoluni in teniplo oppidi SECOTA repositum tamquam custos liegiorum cadauerum. Biria iiiterdum habent in templis hujusmodi idola, iionrLuriquam tria, non plura ; quffl cum obscnro loco sint reposita, horrenda ap parent."
The following plate (xxii.) introduces us to this idol Kiwasa, seated in a sepulchre of the kings, and guarding the repose of the royal dead.
IDOL--WORSHIP.
427
The religion of the Florida tribes is dismissed with the folio-wing brief notice : * " ^Nullam Dei liabent notitiam, neque ullam religionem: quod illis conspicuum est, velrrfci Sol & Luna, illis Deus est. Sacrificos lia bent, quibus valde fidunt: magni euim aunt magi, arioli, tfc dgemonuin invocatoi'es. Funguntur etiam ii sacrifici inedicorum & chirurgorum munere; ejus rei causa semper circumferunt saccum lierbis & medicamentis plenum, ad Eegros curandos, qui plerumque venerea lue laboraiit: nam feminarum & virginum, quas solis filias nuncupant, amoribus sunt admodum dediti."
In plate viii. of tliis " Brevis Narratio," we are in formed tliat tlie Indians venerated as an idol tlie col umn wliicn Hibault liad placed upon a mound to mark the limit of the French empire in the ]STew "World. To this stone they offered the finest fruits, roots, corn, vessels filled with perfumed oils, and. bows and arrows. The column itself was encircled from top to bottom, with wreaths of flowers and branches of choicest trees.
In plate xxxv. w~e are made acquainted with the ceremonies attendant upon the annual offering of a stag to the sun.
Of all the Southern tribes, however, the JSTatchez were probably most addicted to the worship of idols. Pere le Petit' says: " The Natchez have a temple filled with idols. These idols are different figures of men and women for which they have the deepest venera tion." In. another passage he is more explicit: " Their idols are images of men and women made of stone and baked clay, heads and tails of extraordinary serpents,
428
AKTIQUITIES OF TII>J SOUTJIT3RW IXDIA^S.
stuffed owls, pieces of crystal and the jaw-bones of great fishes."
Subsequently, when Father Charlevoix ! visited this temple, its glory had departed--its stone benches were vacant, its idols gone, its altar deserted, and but little left to denote the religious uses to which it had been dedicated save the triangular fire 'watched by the soli tary keeper and slowly burning in honor of the snn. By an old Taenca Indian the Chevalier Tonti 3 was in formed that the natives of that region worshipped the sun and had temples, altars, and priests--" that in the temple there was a fire which, burnt perpetually as the proper emblem of the Sun." To the moon, at certain seasons, oblations were made. Of the temple, the Chevalier lias left us the folio-wing description : " The structure of it was exactly the same -with that of the Prince's house. As to the out-side it is encom passed with a great high "Wall, the space betwixt that and the Temple forming a kind of Coxirt where People may walk. On the top of the Wall are several Pikes to be seen, upon which are stuck the Heads of their own most notorious Criminals, or of their Enemies. On the top of the Frontispiece tliere is a great Knob raised, all covered round with Hair, and above, that an heap of Scalps in form of a Trophy.
" The inside of the Temple is only a J\rave, painted on all sides, at top with all sorts of FigTires ; in the midst of it is an Hearth instead of an Altar, upon which there is continually three great Billets burning, standing up oil end j and two Priests drest in "White Vestments are ever looking after it to make up the Fire
TE5IPIJL OF THE 3TATCHKZ.
429
and supply it. It is round, this that all tlie People come to say their Prayers, with strange kind of Huinmings. Tlie Prayers are tliree times a P>ay ; at Suarise, at Noon, and at Sun-set. They made me take no tice of a sort of Closet cut out of the A\rall, the inside of which was very fine ; I could see only the Koof of it, on tlie top of which there hung a couple of spread Eagles which look'd towards the Sun. I wanted to go into it; but they told me that it was the Taber nacle of their God, and that it was permitted to none but their High Priest to go into it. And I was told that this was the Repository of their "Wealth and Treas ures ; as Pearls, Gold and Silver, precious Stones, and some Goods that came out of JHurope, which they had from their neighbours."
This srm-worship, -with its attendant religious cere monies, wTas not confined to tho tribes who congregated along the banks of the Mississippi, but existed also among1 the Georgia and Florida Indians. Tradition points to a country west of the Mississippi as the orig inal habitat of at least some of the nations composing the Greek Confederacy. "We know that some of the Natchez, abandoning their former seats, joined the Creeks, and it is entirely probable that in doing' so they brought with them their peculiar religious ceremonies, and perpetuated their observance among their new neighbors. Possibly this change of residence may ac count for the introduction of at least some idols or images within the limits of Georgia.
Without further pursuing this inquiry into the re corded observations of the early writers who have en deavored to inform us with regard to the religion of the Southern Indians, it will be perceived that, while we have thus far failed to note any emphatic account
430
ANTIQUITIES OF THE SOUTHERN INDIANS.
declaring the existence of idol-worship among the Georgia tribes, we are certified of tlie fact that idols were seen in the possession of coterminous nations, and that they "were held in superstitious veneration and regarded, in some measure at least, as objects of devo tion. It does appear, however, that they occupied, in the esteem of the natives, a position far inferior to that conceded to the sun or to the Great Spirit, and that they constituted only a sort of religious machinery in the hands of kings, priests, conjurers, and old men,; with which to dignify temples, supplement certain sacred festivals, and operate upon the fears and credu lity of the more ignorant and unthinking masses. One is tempted to regard them rather as conjurers' images, as the private property of priests, as the likenesses of famous dead, and as the potent charms of medicine men, than as the generally acknowledged embodiments of the person and presence of unseen yet recognized divinities.
Although Bolzius, Bartram, Adair, and others, deny either positively or inferentially the existence of idols or images within the limits then occupied by the Georgia Indians, subsequent investigations prove by the discovered presence of the images themselves, that at some time or other idol-worship of some sort was here practised. The ornamented posts, the wooden images, and the questionable figures of men, birds, and animals sketched upon the white walls of the Greek, houses--if any religious significance they possessed-- have long since perished. Next in the order of du rability are small images formed of burnt clay and modelled after the similitude of birds and animals, and of man. (See Plate XXV.) These occur in various parts of the State, and vary in height from
OiAY ABT> STONE lifAGES.
431
three to seven inches. Those which represent the human figure are little more tlmri rude terra-cotta dolls clumsily fashioned. Tlie owl, the wild-cat, and the sun, were favorite subjects for imitation at the hands of the primitive artists. So readily could they have "been made, and so little care was generally bestowed upon their construction, that it may well be questioned whether they amounted, to much more than playthings for children. It may "be, however, that in the reper tory of the priest, the conjurer, and the medicine-man, they possessed greater dignity and were designed for more important purposes.
In a previous chapter we have described several interesting idol-pipes, and have suggested that they were in all likelihood intimately associated with, the religious ceremonies of the aborigines. AVhether they should properly be classed with the simulacra "which we now proceed to consider, we do not confidently affirm or deny. So far as the writer's information ex tends, comparatively few stone idols have been found in Georgia. These occurred in the tipper portions of the State, and chiefly in the valley of the Etowah.
In an old Indian field in Dirt-Town Valley, in Chattooga County, some years since, was ploughed up what may be termed an idol-sanctuary. It was made of a cube of limestone six inches each way. The upper portion or roof consisted of a. quadrangular pyramid, -with a base six inches square, terminating in an apex four inches high, thus giving to the entire ob ject an altitude of ten inches. In one face was an aper ture or doorway, arched at the top, extending almost from the bottom of the structure nearly to the base of the pyramid-shaped top. The interior of this shrine had been carefully excavated, so that its sides and bottom
432
ANTIQUITIES OF THE SOTTTHEBN INDIANS.
were not more tlian lialf an inch in thickness. In front of this arched doorway, and against the opposite wall, a little image was seated upon a pedestal. Only the countenance was visible, the figure being concealed by drapery. On either side was a pedestal similar in form to that upon which the image sat, and about half the size. In the "walls to the right and left of the idol a window had been cut much smaller than, but similar in outline to, the doorway. The whole aifair had been carved out of a solid block of limestone-rock.
In 1860 a stone idol "was found a few miles from Oatoosa Springs. It was about sixteen inches high, and represented a male figure in a sitting posture.
In the possession of Colonel Lewis Tumlin, in 1859, the writer examined an idol which had been ploughed up near the large mound on the Etowah Kiver, upon the plantation of that gentleman. It' was made of a coarse dai-k sandstone, and was twelve inches high. It consisted of a male figure in a sitting posture. The knees "were drawn up almost upon a level with the chin, the hands resting upon and clasp ing either knee. The chin and forehead were retreat ing. The hair was gathered into a knot behind. The face was upturned and the eyes were angular. Unfor tunately, this image was lost or destroyed amid the deso lations consequent upon Sherman's march through Georgia in 1864, but its place has been supplied by another recently found in the same neighborhood. It was ploughed up on Colonel Tumlin's plantation, near the base of the large tumulus" located within the area formed by the moat and the Etowah Kiver, and
'Jones1 "Monumental Remains of Georgia," parti, pp. 108, 109. Savannali, 1861.
2 Ibid., part 1, p. 27, ft seq. Savannah, 1861.
TO\VAH IDOI.
433
is certainly the most interesting idol thus far dis covered in this State. The accompanying front, real', and profile views (see Plate X3TVI.), afford ail intelli gent idea of .the peculiarities of this image. It is a female figure, in a sitting posture. The legs, however, are entirely rudimentary and unformed. Its height is fifteen inches and three-quarters, and its weight thirtythree and a half pounds. Cut. out of a soft taloose rock, originally of a grayish hue, it has "been in time so much discolored that it now presents a ferruginous appearance. Below the navel, and enveloping the but tocks and rudimentary thighs, is a hip-dress, orna mented both on the left side and. behind by rectangu lar, circxilar, and irregular lines. The ears are pierced, and the head is entirely bald. In the centre of the top of the head a hole has been, drilled half an inch ill depth, and five-tenths of an inch in diameter. This probably formed the socket in which some head-orna ment was seated. That ornament, -whatever it was, had fallen out and was lost when, the image was found. Springing from the back of the head and attached at the other end to the back midway between the shoul ders, is a substantial handle by means of which this image could have been securely suspended or safely transported from pla,ce to place. The mammary glands are sharply defined and maidenly in their appearance. The ears, hand, and navel are rudely formed. The impression conveyed is that of a dead, young, flat-head, Indian woman. Unfortunately, the left arm has been broken off, but otherwise this idol is in a state of remarkable preservation. Dr. Berenrlt and Professor liau--to whom the writer exhibited this image--con curred in opinion that this figure bears little resem blance to objects of the same class found in the more
434
ANTIQUITIES OF THE SOUTIIEES INDIAKS.
soutliern parts of this continent. Naked figures are rarely seen there, nearly all of them being clothed, and, generally, highly ornamented, especially about the head, which, in the present instance, is entirely bare and even without ear-ornaments. The features are more like the North American than the Central American Indian, and the remaining; hand is fleshier than is usiially observed among the Central American images. The marks on the back and sides also resem ble more closely the North American pictographs or rock and stone inscriptions than they do the Central American hieroglyphics. "We presume, with the lights now before us, it must be admitted that this is ail Indian image, the handiwork of some nomadic tribe which possessed this region some time during the by gone centuries, and, in turn, "was expelled from the occupancy of this beautiful valley by other and later representatives of the North American type. "Wo are warranted in the assertion that the modern Cherokees 1 disclaimed all share in the erection of the mounds in whose pi'oximity this idol was found. They even went so far as to declare that they possessed not even a tradition of the peoples by whom they "were made, and that their forefathers saw them for the first time in a state of completion -when they occupied the country. They further repudiated the idea that their nation had at any time been addicted either to the manufacture or the -worship of idols. "We may, therefore, safely con clude that this is not a Cherokee idol, certainly of a late date. Where there are no letters, no histories, no inscribed monuments, one wave of human life sweeps over another and the tradition of to-day is swallowed
1 Haywood's " X.itural smd Aboriginal History of Tennessee," p. 226. Kashrille, IS'23.
IETOWAII IDOL.
430
up in the equally frail memory of to-morrow. Under such circumstances emphatically is it true that " one generation passeth away and another generation eometh ; but tlie earth abideth forever. There is no re membrance of former things : neither shall there bo any remembrance of things that are to come with those that shall come after."
"We know not how old this Indian population was. "We cannot even positively assert that it was not autochthonoiTs. "We are ignorant of the distinctive names and characteristics of the various hunter-tribes which may have succeeded each other during the lapsed ages, in the ownership of this soil. As we loolc upon this rude monument, we are not entirely sure that it is emblematic of a past idolatry. It may be the effort of some primitive sculptor to perpetuate in stone the form and. features of some Indian maiden famous in the es teem of her family and tribe.
Various theories may be suggested, fancied analo gies traced and probable origins conjectured, but, after all, the most we cau confidently say with regard to the antiquity of this relic--curious and ancient as it un doubtedly appears--is that it is seemingly older than the handiwork and the superstition of any Indian tribe of which "we have any knowledge as resident upon the beautiful banks of the Etowah.
If object of worship it was, this rude sto-ne image, outliving the generation by \vhich it was fashioned and invested with superhuman attributes, awakened from its long sleep of neglect and desuetude, conveys to us of the present day a true conception of the igno rance and the superstition of that by-gone age, affords physical insight into the condition of the sculptor's art at that remote period, and confirms the past existence
436
ANTIQUITIES OP TIIE SUUTIIURW liSDIAPTS.
of peoples whose names and origin are the subjects only of speculation--whose history is perpetuated sim ply by a few archaic relics which, having1 successfully wrestled with the disintegrating influences of time, re main uricrushed by the tread of another and a statelier civilization. But it is not alone in Georgia that these images are found. Tennessee, above all her sister States, seems to be most prolific of them. In. the be ginning of this century, Mr. Jefferson was presented with two " Indian busts" which were unearthed by some laborers who were excavating along the bant of the Cumberland Elver, near Palmyra. 1 They are described thus : " The human form extends to the middle of the body, and the figures are nearly of the natural size. The lineaments are strongly marked, and sucli as are peculiar to the copper-colored aborigi nal inhabitants of America. It is not known of what materials they arc made : some are of opinion that they have been cut with a chisel or sharp instrument out of stone : others think that they have been moulded or shaped of a soft composition, and afterwards baked. The substance is extremely hard. It has not been as certained whether they are idols or only images of dis tinguished men. It will be an interesting object of research for antiquarians to discover who "were the ancestors of the present Indians capable of executing such a good resemblance of the human head, face, neck, and shoulders."
In his account of the antiquities discovered in some of the Western States, Mr. Caleb Atwater 3 mentions two idols, one found in a tiimulus near JSTashville, Tennessee, and the other dug up on the site of an old
TENNESSEE IMAGES.
4o t
Indian temple in. Natehez, Mississippi. The first was made of clay, peculiar for its fineness, mixed with, gyp sum. The second was of stone. Both are figured in the first volume of the "Transactions and Collections of the American Antiquarian Society." In the writer's collection there is a clay image quite similar in appear ance to Miss Clifford's drawings of the Nashville idol. The accompanying notices of antique idols are ex tracted from Mr. Haywood's " Natural and Aboriginal History of Tennessee."
" Upon the top of a mound at Bledsoe's Lick, in Sunnier County, Tennessee, some years prior to 1823, was ploughed up an image made of sandstone. On one cheek was a mark resembling a wrinkle passing perpendicularly up and down the cheek. On the other cheek were two similar marks. The. breast was that of a female, and prominent. The face was turned obliquely up towards the heavens. The palms of the hands were turned upwards before the face, and at some distance from it, in the same direction that the face was. The knees were drawn, near together : and the feet, with the toes towards the ground, were separated wide enough to admit of the body being seated between them. The attitude seemed to be that of adoration. The head and upper part of the fore head -were represented as covered with a cap, or mitre, or bonnet; from the lower part of which came horizon tally a brim, from the extremities of which the cap ex tended upwards conically. The color of the image was that of a dark infusion of coffee. If the front of the image were placed to the east, the countenance-- obliquely elevated--and the uplifted hands in the same direction, would be towards the meridian sun." *
1 " Natural and Aboriginal History of Tennessee," etc., by Jolm Haywood, pp. 123, 124. NashTille, 1823.
438
ANTIQUITIES OF THE SOUTHERN INDIANS.
Of another image or idol dug up on the farm of Mr. McGilliam on Fall Creek, in Wilson County, Tennessee, Mr. Haywood furnishes the following de scription : " The figure is cut out of a hard rock, of what kind Mr. Ivucker could not determine. It "was designed for a female statue. The legs were not drawn. It only extends a little below the hips. It is fifteen inches long and thick in, proportion. It has a fiat head, broad face, a disproportionately long aquiline nose, low forehead, thick lips, and short neck. The chin and cheek bones are not prominent, but far other wise. On the back of the head is a large projection so shaped as to show, perhaps, the manner of tying and wearing the hair. The nipples are well represented : though the breasts are not sufficiently elevated for a female of maturity. The hands are resting on the hips, the fingers in front, and the arms a-kimbo. Around the back aud above the hips are two parallel lines cut, as is supposed to represent a zone, or belt. The ears project at right angles from the head, with holes through them. It was found a few inches beneath the surface of the earth. No mounds a%~e near, but an extensive burying-ground of apparently great antiquity." '
To the first volume of the " Transactions of the .Amei-icau Ethnological Society," '* Dr. Troost contrib uted drawings of four Tennessee idols. One of them is enshrined in a large cassis flammea, the interior whorls and columella of which had been removed, and the front of the shell cut away so as to permit the en trance and proper location of the image. In these simulacra both sexes are represented. These idols are
" Natural and Aboriginal Hiatory of Tennessee," pp. 1B2, 163. Xiialiville, 1823. '* Pp. 361, 64.
TIESWESSEE EIIJVG-ES.
439
made, some of them of sandstone, and others of a mix ture of clay and shells. All are rude in construction. In the same volume Mr. Schoolcraft, in an article upon the Grave-Creek mound,1 describes and figures a stone idol in a sitting posture, thirteen inches high, which was ploughed up on the farm of a Mr. Taylor some eight miles south of the Grave-Creek Flats.
During his recent investigations, Professor Joseph Jones obtained from the tuimili and valleys of Tennes see several interesting idols both of stone, and of clay mixed with pounded shells. AVithout extending these observations, it may be stated that images of this archaic type have been found also in Kentucky, Vir ginia, South and Worth Carolina, Louisiana, Alabama, and Florida. The scope of the present inquiry does riot lead us to an examination of such as have been ob served in more northern and western localities. The worship of the Priapus probably obtained among some of the Southern Indian nations. In the collection of Dr. Troost were many carefully-carved representations in stone of the male organ of generation. They were found principally within the present limits of the State of Tennessee. But two objects of this sort, so far as our observation extends, have been noted among the relics of the Georgia tribes, and these were about twelve inches long, made of slate. Ill some parts of Alabama, and in Mississippi, similar objects have been exhumed from grave-mounds.
There is another class of objects which commanded the attention, and to all appearances, the veneration and perhaps worship of those ancient peoples. A stone which from some natural cause assumed the
1 "Transactions of the American Ethnological Society," vol. i., p. 408. New York, 1845.
440
ANTIQUITIES OH' THE SOUTHERN INDIANS.
shape of a man or an animal was held in special es teem, and artificial means were sometimes adopted to heighten the fancied resemblance. Such objects were regarded as fit dwelling-places for some manitou or spiritual influence. To Mr. Schoolcraft "we are indebt ed for several illustrations--one of which represents a. natural idol found at the base of a mound in South Carolina.1 From mounds and refuse-piles the writer has obtained relics of this description which doubtless answered some superstitious purpose in the hands of a conjurer, priest, or medicine-man. While the early writers discountenance the idea that idolworship existed among the Greorgia tribes at the period of our first acquaintance with them, remembering the recorded testimony with regard to the religious cere monies, superstitions and practices of other and neigh boring nations who were addicted, at least in some measure, to this sort of adoration, and appreciating the fact that stone idols and clay images have been found not only in portions of this State, but also within the limits of coterminous States, the conclusion seems irre sistible that at some time or other, and among these peoples or those who preceded them in the occupancy Of this region, something like the worship of idols ob tained. Future and more extended observations may enable us more intelligently to comprehend the secrets of the past, and then we will be able to modify, con firm, or reject present conjectures.
1 Scs Squiei-'s '"Antiquities of the State of New York," etc., pp. 171, 172. Buffalo, 1851.
CHAPTER XX.
Pottery.
IT has been truthfully remarked that articles of fictile ware are at once the most fragile and the most enduring of hnmaii monuments. A. piece of common pottery, liable to be shivered to pieces by a slight blow, is more lasting than epitaphs in brass and effigies in bronze. These yield to the varying action of the weather; stone crumbles away, ink fades and paper decays ; but the earthen vase, deposited in some C|uiet but forgotten receptacle, survives the changes of time and, even when broken at the moment of its discovery, affords instruction in its fragments. In their power of traversing accumulated ages and affording glimpses of ancient times and peoples, fictile articles have been compared to the fossils of animals and plants -which reveal to the educated eye the former conditions of our globe. 1
.Perhaps nothing of a physical character more clearly determines the degree of civilization attained by a na tion than the progress made in. the fictile art. In the rudest stages of human existence vessels of some sort are required for the conveyance of water and the prep-
1 "Encyclopedia Britannica," vol. xviii., p. 430, eighth edition.
442
ANTIQUITIES OF THE SOUTHERN INDIANS.
aration of food. Hence, in those remote ages when, we catch indistinct glimpses of man, as an animal, "wrestling witli t-lie lowest wants of his nature and scarce alble to defend. himself against tLe inclement reasons and the attacks of wild "beasts, we find only the meanest forms of domestic utensils, such as gourds, drinking-cups of conch or horn, bark basins, wooden, troughs, skin bags, and coarse earthen pots and pans. The conformation and composition of such primitive pottery indicate the inexperience and awkwardness of the artificers, and convey a decided impression of the barbarity of the race to "which they belonged. As the darkness of a half-clad, nomadic existence is gradually dispelled by the dawning light of civilization, and men. begin to emerge from the savage state, the first step in this development is marked by a change for the better \i\ the ceramic art. The archaic type of pottery is abandoned for forms far more graceful and intellectual, and the crude clay discarded for material more durable and attractive. From its rude beginning to its present stage of picturesque and beautiful development, the pottev's art has always been invested "with peculiar interest and historic value. It may be regarded as the faithful chronicler of man's progress--a fair exponent of the degree of Iris barbarity or civilization, and often the recorder of events and periods -which "would other wise have faded from the recollection of succeeding generations. Hieroglyphic ally impressed upon the sundried bricks of Egypt are the names of a knigly series which, but for these relics, would have irretrievably perished. The sites of ancient Mesopotamia and As syria are traced by means of the cuneiform inscriptions upon the clay bricks of which their proudest edifices were constructed. The Homan. bricks have also borne
HISTORICAL VALUE OF FICTILE WARE.
443
their testimony. Many of them retain the names of the consuls o imperial Rome, while otliers prove that the proud nobility of the Eternal City derived their revenues from the kilns of their Canipanian and Sabine farms. 1
Grecian colonization and its (esthetic influences, remarks Professor Wilson," are traced along the shores of the Mediterranean and the Euxine by beautiful fictile ware and sepulchral pottery. Etiiiria's history is written to a great extent in the same fragile yet en during characters. The footprints of the Roman conqtieror are clearly defined to the utmost limits of im perial dominion, by the like evidence ; and sepulchral pottery is frequently the only conclusive proof which enables the European ethnologist to discriminate be tween the grave of the intruding conqueror and that of the aboriginal occupant of the soil. Apart, therefore, from the exquisite beauty of many remains of fictile art, which confers on them a high intrinsic value, the works of the potter have been minutely studied by the archaeologist and are constantly referred to as historical evidence of the geographical limits of ancient empires.
Few peoples, how degraded soever, have failed to bequeath some specimens of pottery--crude and mis shapen though they be--to rescue the fact of their former existence from utter oblivion. The absence of pottery in the Reindeer period in France furnishes a decided exception, and affords proof alike of the great antiquity of the cave-dwellers of Dordogne and of the very low state of their civilization,3
In Europe, where prehistoric archsoology may be
444
ANTIQUITIES OF THE SOUTHEBN INDIANS.
divided into four great epochs--the PalEeolitliic and the Neolithic periods, the Bronze and the Iron ages-- careful study has been bestowed upon the peculiar characteristics of the pottery of each period. The con clusion to which this examination leads is this, that while the potter's wheel was probably unknown ill both the Stone and Bronze epochs, the material of which the Stone-age pottery is composed is rougher than that which was used during the Bronze period. The ornaments of the two periods show, also, says Sir John Lubbock,* a great contrast. In the Stone age they consist of impressions made by the nail or the fin ger, and sometimes by a cord, twisted round the soft clay. The lines are all straight, or, if curved, are very irregular and badly drawn. In the Bronze age all the patterns present in the Stone age are continued, but in addition we find circles arid spirals ; while imitations of animals and plants are characteristic of the Iron age.
In North America, where we have almost exclu sively a Stone age distinguished, by relics more varied than those of perhaps any other quarter of the globe, the art of pottery attained a considerable degree of per fection. The ornamentation is as diversified and com prehensive as tliat of all the ancient epochs of Europe combined. The manufacture of fictile articles for do mestic use, devotion, and ornament, seems to have been carried on by most of the Indian tribes, from time im memorial. The Southern Indians excelled in the ceramic art, special care having been bestowed upon the selection and preparation of their clays, and 110 little taste displayed both in the shape and ornamenta tion of their vessels. The use of these frail utensils
1 " Prehistoric Times," p. 16, second edition. London, 1S69.
POTTERY OF TBVE FLOKmA INDIANS.
445
was, however, at an early period superseded by the employment of more serviceable articles obtained from the whites, and the fabrication of pottery was, with but few exceptions, speedily abandoned whenever am ple opportunity was afforded for the purchase of Euro pean copper kettles, iron pots, and tin-ware. This fact increases our interest in these perishable relics, and causes us to cherish very tenderly all specimens of this character.
At the period of oTir first acquaintance with the Southern Indians, the fabrication and use of earthen vessels were very general. The Fidalgo of Elvas' pays high compliment to the pottery of the regioja when he describes it as " little differing from that of Estreinoz or Montemor." He records the circumstance that the natives " had great store of walnut oil--clear as butter, and of a good taste--and of the honey of bees pre served in pots,"
It would appear from Cabeea tie Vaea's account 2 that some of the Southern tribes were either ignorant or neglectful of the potter's art. Of such he Avrites : " Their method of cooking is so new, that for its strangeness, I desire to speak of it; thus it may be seen and remarked how curious and diversified are the con trivances and ingenuity of the human family. Wot having discovered the use of pipkins to boil what they would eat, they fill the half of a large calabash with water, and throw on the iire many stones of such as are most convenient and readily take the heat. When hot, they are taken up with tongs of sticks and dropped into
446
ANTIQUITIES OF TUB SOUTHERN IKDIANS.
the calabash, until the water in it boils from the fervor of tlie stones. Then whatever is to bo cooked is put in, and, until it is done they continue taking out cooled stones and throwing in hot ones. Thus they boil their food." In the " Brevis Narratio "' of Le Moyne de Morgues, we have several illustrations pur porting to exhibit the forms of pottery in general use among the Florida Indians.
The Chevalier Tonti, in his general description of the Louisiana Indians, uses the following language : " They have Cellars or rather Holes to preserve their Corn, their Wood and other .Provisions ; but all their Kitchin Utensils consists in some few pieces of Earth en-Ware which they make with Clay, and harden it with the Dung of Bulls." "
Father Ilennepiu s asserts that before the arrival of Europeans in North America " both the Northern and Southern Salvages made use of and do to this day nse Earthen Pots, especially such as have no Commerce with the Europeans from whom they may procure Kettels and other jVToveables."
During Lieutenant Timberlake's sojourn among the Cherokees he observed that they nsed two sorts of clay from which they made excellent vessels capable of resisting the greatest heat. At a physic-dance in the town-house he saw a clay pot, set on the tire, capa ble of containing twenty gallons. 4
Speaking of the same Indians Adair" asserts that
1 Flates riii., xi., xx., xxviii., xxix. Francoforti ad Mcemim. De Bi-y, anno 1591.
a " An Account of Monsieur de la Salle's Last Expedition," etc., p. 12. Lon don, 1(108.
3 " Continuation of the Isrevv Discovery of a Vast Country in America," etc., p. 102. London, 1698.
> "Memoirs," etc., pp. 62, 77. London, 1Y65. * " History of the American Indians," p. 424. London, 1V73.
POTTKHY OF THE LOUISIANA IXDIAK3.
4:i7
in his day they made " earthen pots of very different sizes so as to contain from two to ten gallons ; large pitchers to carry water, bowls, dishes, platters, basoiis, and a prodigious number of other vessels of such anti quated forms as would be tedious to describe and im possible to name. Their method of glazing them is, they place them over a large fire of smoky pitch-pine which -makes them smooth, black and firm. Their lands abound with proper clay for that use; and even with porcelain, as has been proved, by experiment."
Loskiel * tells us that the Delawares and Iroquois . had pots and boilers made of clay mixed with pounded sea-shells, and burnt so hard that they were black throughout; and Joutel affirms that the Indians inhabiting the banks of the Mississippi were very skil ful at making earthen vessels wherein they boiled their flesh, or roots, or sagamise.2
In commenting upon the customs of the Louisiana tribes, Du Pratz 3 writes : " To prepare their maiz for food and likewise their venison and game there -was necessity for dressing them over the fire, ancl for this purpose they bethought themselves of earthen ware which is made by the women who not only form the vessel, but dig up and mix the clay. In this they are tolerable artists: they make kettles of an extraordi nary size, pitchers with a small opening, gallon bottles with long necks, pots or pitchers for their bear oil which will hold forty pints ; lastly, large and small plates in the French fashion. I had some made out of curiosity upon the model of my delf ware, which
1 " History of the Mission of the United Brethren," etc., part 1, p. 54. Lon don, 1794.
1, p. 149. New York, 1846. B " History of Louisiana,' 1 p. 360. London, 1774.
448
ANTIQUITIES OF THE SOLTCHEKJr INDIANS,
were a very pretty rod." The Idtcheii utensils of the Alabama Indians consisted of dishes and pots of earthen-ware, and deep wooden dislies. " They made cups of calebashes, and spoons of the horns of wild oxen, which they cut through the middle and form into the proper shape by means of fire." ' Ilaywood " mentions the existence in the mounds of Tennessee of fragments of pottery composed of clay and pounded cockle-shells.
It was liawaoii's impression that the earthen pots round buried, and at the foot of banks whence the water had washed them, wore of a sort different from those in use by the Carolina Indians when, he so journed among them. He asserts that the ancient pottery was "thicker, of another shape and composi tion, and nearly resembled the urns of the ancient Komaiis." 3
In plate xv. of Harlot's " Virginia," -we are advised of the method in which the natives seethed " their meate in earthen pottes." Says the translator: "Their woemen know how to make earthen vessells with special Cunninge, and that so large and line, that ouv potters with thoye wheles can make noe better; ant then Kemoue them from place to place as easelye as we can doe our brassen. kettles. After they haue set them iippon an heape of erthe to stay them from falliuge, they putt wood vnder, which, being kyndled, one of them taketh great care that the f'yre Imrne ecruallye lioimde abowt. They or their woemen fill the vessel with water, and then putt they in finite, flesh, and fish, and lett all boyle together like a gallie-
1 Bossn's < : Travels through Louisiana," vol. i., p. 224. London, 1771. 3 "Natural and Aboriginal History of Tennessee," p. 139. Nashville, 1823. "History c,f Carolina," pp. 278, 279. Eepriut. Kaleigli, I860.
MANUFACTURE OF CLAY UTENSILS.
449
maufrye, which tile Spaniarde call olla podrida. Then they putte yt out into disclies, and sett before the compaiiye, and then they make good clieere to gether." '
We conclude these citations in support of the fact that the Southern Indians at the time of primal con tact between them and the whites were almost uni versally cognizant of and practising the' potter's art, by an observation of that intelligent and entertainingtraveller, AVilliam Bartram,' to whom we are indebted for so much valuable information respecting the Geor gia tribes in 1773 : " As to mechanic arts or manufac tures, at present they have scarcely any thing worth observation, since they are supplied with necessaries, conveniences, and even superfluities by the white traders. The men perform nothing except erecting their mean, habitations, forming their canoes, stono pipes, tambour, eagle's tail or standard, and some other trifling matters ; foi1 Avar and hunting are their principal employments. The women are more vigi lant, and turn their attention to various manual em ployments: they make all their pottery or earthen ware, inocasins, spin and weave the curious belts and diadems for men, fabricate lace, fringe, embroider and decorate their apparel," etc., etc.
The statement, therefore, is historically correct that until, through their intercourse with the early explor ers and first settlers, the Southern Indians became con vinced of the superiority of the copper and iron kettles and articles of crockery then introduced and of their
450
ANTIQCTTIES Op THE SOUTHERN INDIAKS.
ability to possess them, they adhered to their primi tive manufacture of clay utensils of various forms. With the general introduction of these more durable articles of European, construction dates the decline of the ceramic art among the rSTortli American tribes.. That decadence was more or less rapid as the inter course between the races became partially or perma nently established ; and to such an extent has it pro gressed that, in the language of Prof. Rau, at the pres ent time this aboriginal art may be considered as al most if not entirely extinct among the tribes still in habiting the territory of the United States, excepting some in ]STew Mexico and Arizona who have not yet abandoned the manufacture of earthen-ware.
In York County, South Carolina, dwell some sixty survivors of the once powerful Catawba nation. By them the falsification of iictile articles lias not been wholly discontinued. This is clone, however, rather with a view to satisfying, at a good price, the demands of strangers who make frequent application for their wares, than in perpetuation of the ceramic art as it once existed among them.
The pottery of the Southern Indians is superior to that manufactured by Northern tribes. It is more varied in form, symmetiical in shape, excellent in composition, and diversified in ornamentation. The abundance of choice clay, a climate salubrious the year round, the presence of fish and game in plenty, and the fact that ISTature spontaneously gratified many wants--combined with the general dissemination of art ideas apparently derived from the Natchez--afforded ample leisure and facilities for the careful fabrication of fictile ware and tended to develop a degree of taste and sliill which not infrequently challenges our adrnira-
MANCFACTUKE OF POTTERY.
451
tion. The presence of sherds all over the cultivated fields attests the numbers of clay vessels which wrere everywhere iu use among- tlie aborigines of Georgia. Especially do these fragments abound upon tlie sites of their villages and at the principal bluffs along- the coast and water-courses whither, in ancient times, they resorted for the purposes of fishing and hunting. The refuse-piles are here filled with broken clay utensils thrown aside as they perished with the using. They form an important element in the debris of the en campment. Seldom are entire vessels found except in mounds and graves. Even here it is a difficult matter to secure specimens wholly free from blemish. Friable in its cliaract(3r, this pottery was liable to disintegration. Under the most favorable circumstances, when securely deposited iu tumuli, the moisture of the soil and the weight of the superincumbent mass of earth in many instances caused the burial-urns and cooking-utensils to crack or fall to pieces. The sepulchral shell-niomids and the dry sandy tumuli of the coast were most con ducive to the preservation of these frail articles. From them the best specimens have been taken.
The material employed by the Greorgia Indians in the manufacture of their pottery was red, blue, yellow, and dark-colored clay. It was often used without the admixture of any foreign substance ; but in many cases this clay was tempered, mixed and kneaded with powdered shells, gravel, or pulverized mica. Experi ence taught these primitive artificers that such a com position imparted greater consistency to the mass and rendered it more capable of resisting the action of fire.
Sometimes--as in the case of flat-bottomed vessels intended as receptacles for pounded maize--this pot tery \vas only sun-dried, but generally the utensil was
452
ANTIQUITIES Of THE SOUTHKBK ISDIASS.
subjected to a hardening process by fire. Tlie appli cation of lieat to the interior of the vessel was occa sionally so intense as to cause a partial fusion of the inner particles. This pottery appears to have been made fay hand, although, so accurate are its outlines, so homogeneous its composition, and so regular the thicknesses of the Avails, that we often wonder how it could have l>een so skilfully formed without the aid of the potter's wheel. This earthenware was manufac tured by the natives in almost every part of the State. Traces of the pits whence they dug clay, are still ex tant. Scattered around are fragments of pottery, masses of clay evidently intended for use, and the re mains of former fires. Localities where these potters plied their trade may, to this day, be clearly noted on the coast, in the valleys of JLittle-Shoulder-Bone Creek, of the Etowah, Oostenaula, and Chattahoochee Rivers, and elsewhere. In Bibb and Cass Counties rude clay. hearths with elevated sides have been imearthed, which, from their form and the quantities of sherds in their vicinity, suggest the belief that they were crude kilns for baking pottery. Professor Ran, in his interest ing article on " Indian lottery," furnishes a valuable ac count of some localities on the left bank of Cahokia Creek, in the American Bottom, where the manufacture of earthen-ware had been carried on by the Cabokia Indians. " In some of the Southern States," remark Messrs. Squier and Davis,' "it is said the kilns in which the ancient pottery was baked are now occa sionally to be met with. Some are represented still to contain the ware, partially burned, and retaining the rinds of the gourds, etc., over which they were mod elled, and which had not been entirely removed by
1 " Ancient Monuments of the Mississippi Valley," p. 195. "Washington, 1848.
POTTERY--KILNS.
453
the fire." " In Panola County," says Mr. R. Morris, in a private letter, " are found great numbers of what are termed pottery-Jcilns, in which are masses of vitri fied matter frequently in tlie form of rude bricks meas uring twelve inclies in lengtli by ten ill breadtli."
In the Etowah Valley -- a region of all others in Georgia most rich in monuments--ovens, rudely con structed of water-worn stones, have been discovered, witli circular paved floorings indicating the long-con tinued presence of hot fires. Those which the writer examined were in ruins, but seemed to have been about five feet in diameter. The impression created by these remains and their surroundings was, that they were intended for and used as kilns for baking pottery.
Observing for a moment the general characteristics of the pottery found in Georgia, we will note that the Avails or sides of the vessels vary iu thickness from the eighth to the half of an inch. Some of tile largest sort are thicker still, their bottoms being reenforced to in sure additional strength. In size there is every varie ty, from the little poculum capable of holding scarce a pint, to the large pot or flat-bottomed jar whose con tents may be calculated by the gallon. Most of the vessels belonging to what may be termed the archaic type, are but slightly ornamented ; many of them not at all. The same may be affirmed with regard to the coarser jars designed as receptacles for pounded maize, bear-oil, walnut-oil, and honey. The rinis of not a few of the larger vessels curve outward, so as to allow a vine or cord to pass round and under the projection, and thus enable them to be suspended over the fire. Others lave strong ears, by means of which suspension could have been accomplished with greater facility.
454
ANTIQUITIES OF THE SOOTHBRN INDIANS.
Others still lack both curved rim and ears ; and, with tlieir rounded bottoms, must have been kept in an up right position through the intervention of clay rings placed beneath, or by being bolstered tip with stones, fagots, or sand. One of the best specimens of ceramic art we have seen within the geographical limits of Georgia, is the burial-urn represented by Fig. 1, Plate XXVII.
It is fifteen inches and a half in height, nine inches in diameter in the widest part, and ten inches and a q/uarter across the top. The graceful outline and gen eral symmetry of this vase arrest our attention. It was apparently made with the assistance of a rush or wicker basket, as its entire exterior surface is covered by impressions left by the rushes or osier-twigs upon the clay while in a plastic state. "We know that some of the [North _A_tnerican tribes adopted the custom of modelling their vessels in baskets prepared for that purpose. Either this method was used in the present case, or else the potter, with no little skill and pa tience, imprinte'd these ornamental lines "while the ves sel was still soft, by means of a cord or instrument of some sort. The lines are impressed, not carved. The circular ornamentation--running parallel with and half an inch distant from the rim--was doubtless made with the end of a hollow reed or bone. The hard cane abundant everywhere in the swamps of Southern Georgia and generally used by the Indians for arrows, might well have been, employed for this purpose. The interior is. quite smooth. This urn was fashioned of the clay common to the neighborhood in which it was found. In its composition there is an admixture of gravel, and, to a limited extent, of pow dered shells. In itself considered, it is a creditable
example of the skill of the primitive potter. It pos sesses, however, an individual history which invests it with additional interest.
This burial-Tim was found in a small shell-mound on the Colonel's Island,' in Liberty Comity. It -was in an upright position and its rim was about eighteen inches below the surface. This little tumulus was evi dently very old; and, although the ploughshare had not torn it asunder, the changing seasons and the mer ciless winds and rains had sadly wasted it. But for the quantities of stout oyster-shells which, entered into its composition it would long since have been oblit erated by these disintegrating influences. The remark able state of preservation in which, this vase appears is accounted for when, we are made acquainted with the fact that it was guarded or enclosed by two exte rior earthen vessels of ruder construction and thicker walls. Covering the top of the outer vessel and closely fitting, -was a substantial lid or cap of baked clay, made for the purpose. The exterior and middle receptacles were so much softened and. impaired by the moisture of the sand and shells that they crumbled into fragments in the effort to remove them and could not be restored. It was -with difficulty that the in nermost urn could be lifted from its position. Expos ure to the sun, however, soon caused it to harden. Within this smallest and enclosed vessel, thus pro tected, were the bones of a young child. They had wellnigh returned to the mother-dust from "which they
1 In l*?^, William Bai-tram, while on a visit to this island, observed amon;; the shells of a conical mound, and about its centre, the rim of an earthen pot which he carefully removed, drawing it out almost whole. '= This pot," he says, "was curiously wrought all over the outside, representing basket-work, and was undoubtedly esteemed a very ingenious performance by the people at the age of its construction."--(" Travels through North and South Carolina, Georgia," etc., p. G. London, 1792.)
456
ANTIQUITIES OF THE SOUTHERN" INDIANS.
sprang. No relics, save these funeral-vases, were found in this mound. This fact suggests two thoughts --one, that the tumulus was erected solely in honor of this infant, and consequently that it must have been the offspring of some noted personage of the tribe; the other, that it was too young to have come into the ownership of any articles except such as must have been very perishable in their character.
Placed here perchance by the "wife of the chieftain --certainly by an affectionate mother--with the fond hope that this clay coffin, in all likelihood her own handiwork, would shield the tender form of the babe she loved so "well from the chilling damp and the re morseless decay of the lonely grave, this funeral-vase affords an affecting illustration of that sincere natural attachment which leads even the uncivilized parent to "wrestle with death for the preservation of her buried child. Three other instances of similar inhumations have chanced within the writer's observation, all of them occurring in mounds on the coast. It will be remarked that this sepulchral-urn is not xvnlike those described by Mr. Atwater and figured 011 pages 227 and 229 in the first volume of the " Arcliaeologia Americana." Burial-vases enclosing human bones have occasionally been found in the grave-mounds of Ten nessee, Alabama, Florida, Mississippi, and South Caro lina. In ancient Greece it "was customary to deposit the ashes or bones of the dead in a cinerary of baked clay, bronze, or gold, and recent investigations show that this method of protecting the dust of the departed was not confined to the limits of classic Hellas. The vessel (Fig. 2, Plate XXVII.) taken from an earthmound near Sparta, in Hancock County, is fourteen inches high and rather more than fourteen inches in
VARIOUS FORMS OF AKCTENT POTTEEY.
457
diameter. Near the rim we have a repetition of the circular or bead ornamentation noticed on the fourialvase. The ornamentation of the entire outer surface is so varied and elaborate that we are somewhat at a loss to understand precisely how it was done. If this pot was moulded in a basket, the pattern of the en closing wicker-woi*k was unusually elaborate and ar tistic. _A-S in the case of the sepulchral lira, all these impressions were formed while the clay was still soft. There are no indications of the use of a sharp-pointed implement as in vessel Number 3, Plate XXVII., where all the lines and figures were carved after the clay had become hard.
Fig. 4, Plate XXVII., may be regarded as typical of a numerous class of flat-bottomed jars designed, as has already been intimated, as receptacles for various articles, such as pounded maize, bear-oil, walnut-oil, honey, etc. It is entirely plain both within and with out, quite smooth, and measures rather more thaa eight inches in height and nine inches in diameter. The dark clay of which it is composed was tempered with powdered shells and mica.
Figs. 5, 6, and 7, Plate XXVII., are accurate deline ations of pots with ears, while Fig. 8, in general out line, assimilates very closely to the small iron pot of the present day. The addition of legs was by no means usual. Figs. 9 and 10 of the same plate acquaint us with the shapes of the ordinary clay bowls in common use among the primitive peoples of this region. In Figs. 1 and 2, Plate XXVIII., we observe the forms of the wide-necked jars. The vessels delineated in Figs. 3 and 4, of the same Plate, were taken from an ancient burial-ground in the Mississippi Valley, near Shreveport, while those represented by the remaining Figures
458
ANTIQUITIES OF THE SOUTHERN INDIANS.
in this Plate were found by Professor Joseph Jones, in the stone-graves and mounds of Tennessee. "Without multiplying these examples, the illustrations already furnished advise us of the prevailing types of this South ern pottery. .All these vessels were taken from gravemounds. Animal-shaped and face-vases occasionally oc cur. Of the latter kind the well-remembered and oftdescribed triune vase is a striking illustration. Pro fessor C. Eau is now preparing a monograph upon facevases which will prove both curious and interesting.
Upon an examination of this pottery and the many sherds which everywhere abound (some of which are figured in Plate XXIX.), we are led to believe that the ornamentation was compassed in one or the other of the following ways :
I. By modelling the vessel inside of a net-work, rust-basket, or frame made of twigs or split cane, or within a gourd, or over blocks of wood or forms of dried clay. It seems, moreover, from the delicacy of some of the impressions, that a sort of cloth must have been first spread against the sides of the enclosing basket or framework before the clay was put in and pressed against it. Perhaps in some instances the in terior walls of the gourd may have been carved so as to leave raised figures and lines upon the vessel moulded "within it.
II. By shaping the kneaded clay into the desired form, with the liand, leaving the outer surface smooth ; and, when the pot was dry, with a sharp flint-flake or bone carving straight, curved, and zigzag lines with greater or less uniformity according to the care, pa tience, and skill of the artificer.
IIT. The circular and semicircular depressions-- -with or without elevated centres--could have been
pi
s.
I' v'lWj
VARIOUS METHODS OF ORNAMENTATION.
459
made by means of a hollow reed cut off at or near a joint, as might best indicate the artist's present fancy. It is not improbable that some of the indentations formed while the clay was still in a plastic state, were clone with the finger-nail, which the Indians, in some cases and for certain purposes, permitted to grow very long.' Lilies were impressad with the aid of a thong, while the more complicated figures may have been perpetuated with the assistance of a wooden, or soapstone die iu which the desired pattern was cut. Kepeated applications of the same die to all the exterior portions of the vessel gave a uniform ornamentation. The use of several dies of different designs materially enhanced the. variety.
IV. Frequently raised mouldings near the rims, and elevated ornaments were added while the vessel was still soft, and when the adhesion of these new parts could be readily compassed.
V. The sides of the vessels were sometimes beau tified by the insertion of diamond and square-shaped, parallelogrammic, and circular pieces of inica and shell. Over the edges of these inserted or impressed orna ments the clay was slightly curved, so that when, the ware was thoroughly dry these pieces of mica and shell remained permanently embedded. A beautiful driiikiiig-cup ornamented ill this way was unearthed by a freshet, wdiich, overflowing the banks of Savannah River, cut a channel through an ancient burial-ground near the confluence of great Kaokee Creek and that stream.
VI. The ornamentation of this earthen-ware was further accomplished by means of red, blue, and black pigment.
1 Lawson's " Carolina," p. 284. Raleigh reprint, 1860.
460 AjrriQxrrrrEs OF THE SOUTHERN INDIANS.
AVhen completed tlie newly-formed vessel was either exposed in the sun, baked in a kiln or open fire, or inverted over burning coals of some hard wood, such as oak or hickory, piled up so as to fill as nearly as possible the whole interior. In the manner last mentioned was the baking process often conducted, the bed of coals being at intervals renewed and ar ranged in conical form so as to distribute the heat equally to every part of the pot. So intense at times was the heat employed, that the vessel glowed and a fusion of the particles on the inner surface occurred. When sufficiently baked, the vessel was allowed to cool gradually, in its hardened condition permanently retaining the impressions which had at first been made upon its plastic form.
Upon the manufacture of the ordinary cooking and domestic utensils comparatively little labor was ex pended. They were, however, substantially made, and answered well, both in shape and durability, the wants of this primitive period. From some of the sepulchral tumuli and refuse-piles, plates of baked clay, usually about six inches long, four inches wide, and half an inch thick, have been taken. They were used either as plates or rude baking-pans. Clay pans, with numerous holes pierced through their bottoms, thereby converting them into convenient strainers, have also been found.
In Cherokee Georgia and AJabama frequent use was made of pot-stone or soapstone for the manufac ture of vessels of the largest size. Some of this kind have been exhumed fully three feet in diameter and eighteen inches deep. The walls were an inch and upward in thickness,'
1 Professor Joseph Jones 1ms a ve nearly two hundred pounds ; and one permit an adult to sit and bathe in it.
MANUFACTURE OP POTTERY.
401
We here omit a description of images, pipes, "beads, and other articles of clay, as they have been noticed iu another connection.
As the methods adopted by the various American nations in the manufacture of their earthen-ware -were probably quite similar, in addition to the extracts already given, it is deemed proper to present the fol lowing accounts by eye-witnesses as throwing addi tional light upon this subject.
In his history of the manners and customs of sev eral Indian tribes west of the Mississippi, Hunter ob serves : * "In manufacturing their pottery for cooking and domestic purposes, they collect tough clay, beat it into powder, temper it "with water, and then spread it over blocks of "wood which have been formed into shapes to suit their convenience or fancy. When suffi ciently dried, they are removed from the moulds, placed in proper situations and burned to a hardness suitable to their intended uses. Another method practised by them is to coat the inner sxrrface of baskets made of rushes or willows, -with clay to any required thickness, and, when dry, to burn them a,s above described.
" In this way they construct large, handsome, and tolerably durable ware; though latterly, with such tribes as have much intercourse with the whites, it is not much used, because of the substitution of cast-iron ware in its stead.
a W^hen these vessels are large, as is the case for the manufacture of sugar, they are suspended by grape vines, which, wherever exposed to the fire, are con stantly kept covered with moist clay.
" Sometimes, however, the rims are made strong and project a little inwardly, quite round the vessels, so as
1 "Memoirs of a Captivity," etc., p. 288. Loiiflon, 182S.
402
ANTIQUITIES OP THE SOUTHERN INDIANS.
to admit of their Toeing sustained by flattened pieces of wood, slid underneath these projections and extend ing across their centres."
The Mandaris are reported by Mr. Catlin 1 to have fabricated their pottery ill the following manner: " Earthen dishes or bowls are a familiar part of the culinary furniture of every Mandan lodge, and are manufactured by the women of this tribe in great quantities, and modelled into a thousand forms and tastes. They are made by trie hands of the womenfrom a tough, black clay, and baked in kilns, which are made for the purpose, and are nearly equal in hard ness to our own. manufacture of pottery ; though they have not yet got the art of glazing, which would be to them a most valuable secret. They make them so strong and serviceable, however, that they hang them over the fire as w~e do our iron pots, and boil their meat in them with, perfect success. I have seen some few specimens of such manufacture, which have been dug up in Indian mounds and tombs in the Southern and Middle States, placed in our Eastern museums, and looked upon as a great wonder, when here this novelty is at once done away with, and the whole mys tery ; where women can be seen handling and using them by hundreds, and they can be seen every day in the summer also, moulding them into many fanciful forms and passing them through the kiln where they are hardened."
The most minute account is that furnished by Drmiont, who, in describing the customs of the Louisi ana Indians, states that, " after having amassed the proper kind of clay and carefully cleaned it, the Indian
1 " Illustrations of the Manners, diatoms, and Conditions of the Novtn Amer ican Indians," etc., vol. i., p. 116. London, 1848.
MASUFACTUKE OJT POTTEBY.
463
women take shells which tliey pound aiicl reduce to a fine powder; tliey mix this powder with the clay, and having poured some water on the mass, they knead it with, their hands and feet and make it into a paste, of which they form rolls, six or seven feet long, and of a thickness suitable to their purpose. If they intend to fashion a plate or a vase, they take hold of one of these rolls by the end, and, fixing here with the tlmmb of the left hand the centre of the vessel they are about to make, they turn the roll with astonishing quickness aroniid this centre, describing a spiral line; now and then they dip their fingers into water and smooth with the right hand the inner and outer surface of the vase they intend to fashion, which would become ruffled or undulated without that manipulation. In this manner \they make all sorts of earthen vessels, plates, dishes, bowls, pots, and jars, some of which hold from forty to fifty pints. The burning of this pottery does not cause them much trouble. Having dried it in the shade, they kindle a large fire, and, when they have a sufficient quantity of embers, they clean a space in the middle where they deposit their vessels and cover them with charcoal. Thus they bake their earthen-ware, which can now be exposed to the fire, and possesses as much durability as ours. Its solidity is doubtless to be attributed to the pulverized shells which the women
mix with the day" 1 The ceramic art is no longer practised by the Indian
within the limits of Georgia. Upon the removal of the Creeks and Cherokees, the last representatives of the red race departed from the beautiful valleys, the noble mountains, and luxuriant forests of this Empire State
'Dumont, " Memoii'CS Historiques sur la Louisiiuic," torre ii., p. 271, et scg.
Paris, 1153.
464
ANTIQUITIES O3? THE SOTJTIIEBH INDIANS.
of the South. Even before the establishment of Ogletliorpe's colony at Savannah, there had occurred a by no means partial introduction of metallic vessels of Euro pean manufacture. These were furnished by traders who swarmed into the Indian country from Carolina arid the Spanish settlements in Florida. Upon the gen eral distribution of these more durable utensils, the fabrication of fictile ware gradually subsided and was at last entirely abandoned. To the industry and skill of the Indian women of those early days are we mainly indebted for these interesting relics of the past.
Roguet l advanced the idea that the -way in which pottery came to be made was this: primitive peoples at first daubed with clay such combustible vessels as cocoa-nut shells, to protect them from the action of fire. It was found before long that the clay itself, when hardened, would retain its shape and answer the pur poses of the vessels it was designed to enclose. Thus the idea of fictile ware was conceived and from time to time developed. The observations of Captain Genneville and others tend to corroborate this notion; and it may be that the early efforts of the Southern Indians in the ceramic art were confined to covering gourds with clay so as to use them for culinary purposes.
Although calabashes were long ago abandoned as unsuitable for heating water and boiling maize, the shape of many of the terra-cotta vessels of an antique type would seem to have been suggested by them.
Aside from the disintegrating influences of time and moisture, the casualties of use and accident, the operation of inherent decay and the wanton destruction of many of these frail vessels at the hands of the care-
1 See Tylor's " Researches into the Early History of Mankind," etc., second edition, p. 273. London, 1870.
CHEMATION.
465
less and the unlearned, the Southern Indians, in observ ance of a custom which obtained among some ancient tribes, doomed to destruction quantities of their pot tery. It will be remembered, that these primitive peoples', especially along the coast of Georgia, fre quently burned their dead and with them food-vessels, drinking-cups, pots, flagons, ornaments, utensils, and articles, the property of the deceased. The practice of reserving the skeletons until they had accumulated, sufficiently to warrant a general inhumation was main tained among the Creeks, the Choctaws, and other Southern nations within, the historic period. It was no easy task, as "we have already observed, for the aborigines, with their limited means, to erect a tumu lus. Hence, by an arrangement of this sort, the com bined labors of the many could be secured in compass ing the elevation of gi'ave-mounds above the accumu lated dead of village or tribe. Possibly, cremation "was resorted to in order that the toil of mound-building might be diminished. Cremation, however, was by no means universal even iu districts where the dead were frequently burned. "Why these funeral customs should have thus varied in prescribed localities, we do not fully understand. Compared with each other these sepulchral tumuli differ materially in their ages, and we can only repeat what we have already suggested in explanation, that in the history of the nomadic peoples who for centuries possessed this region, one wave of human life may have swept over the other, each perpetuating its peculiar funeral-rites, and leaving in silent companionship mound-tombs similar in gen eral aspect and yet possessing internal indicia which intimate that they are the creations of different hands,
466
ANTIQUITIES OP THE SOUTHEBN INDIANS.
the offspring of varying customs--all designed, how ever, to honor the memory of the departed.
But a few weeks since the writer opened two gravemounds, not more than forty yards apart, in the midst of an ancient-Tburial-ground on the Georgia coast. In the first, the skeletons had been disposed in an horizon tal position and the smell of fire had not passed upon them. In the other, after having been collected in a circle twenty feet in diameter, with all their articles of property about them, the dead, to the number of perhaps thirty, had been consumed in the flames. Charred fragments of -wood and bone, broken pieces of pottery, cracked stone implements, and burnt earth, abtindantly testified how complete had been the crema tion. Here was a total demolition of numerous clay vessels owned by the deceased and given to the flames with the skeletons prior to the inhumation. Bushels of fragments might have been gathered, but not a vessel remained in its entirety to reward the investi gation.
Upon the burning -n-vpai the Greeks cast perfumes and oils, but the beautiful vases and the property of the deceased were claimed by the living. The South ern Indian gave to one common funeral-flame the skeleton, and all the possessions of the departed.
CHAPTER XXI.
The Use of Foarls as Ornaments among the Southern Indians.
IN tlie concession made by the King of Spain to Hernando de Soto of the government of Cuba and conquest of Florida, with the title of A-delantado, onefifth of all the gold and silver, stones and pearls, won in battle or on entering towns, or obtained by barter with the Indians, was reserved to the crown. It was further stipulated that the "gold and silver, stones, pearls, and other things which might be found and taken as well in the graves, sepulchres, ocues or temples of the Indians" as in other places where they were ac customed to offer sacrifices to idols, or in other con cealed religious precincts or buried houses, or in any other public place," should be equally divided between the king and the party making the discovery.
From the special mention made of them in this royal reservation, it is evident that among the valu able trophies of the expedition precious pearls were confidently anticipated. That the Spaniards were not entirely disappointed in this expectation the early narratives abundantly testify. These relations estab lish the fact--and that beyond all controversy--that the use of the pearl as an ornament, among the Indians
468
ANTIQUITIES OF THE SOUTHERN- INDIANS.
of Florida and of the South -was by no means infre quent. A reference to some of these accounts--afford ing as they do the earliest information \ve possess 011 this subject--may prove interesting.
Near the bay of Espiritu Santo, in Florida, the followers of De Soto chanced upon the town of an In dian chief--Ucita, by name. His house stood near the beach, upon an artificial mound. At the other end of the towti was a temple, on the top of which perched a wooden fowl with gilded eyes. Within these eyes, says the historian, were found pearls such as the Indians greatly value, piercing them for beads and stringing them to wear about their necks and wrists.
^Vhen the Indian queen welcomed the Spanish adventurer to the hospitalities of Cutifachiqui, she drew from over her head a long string of pearls, and, throwing it around his neck, exchanged with him gracious words of friendship and courtesy. Observing that the Christians valued these pearls, the cacica told the governor that if he would order some sepulchres, which -were in the village, to be searched he would find many; and, if he chose to send to those which were in the uninhabited towns, he might load all his horses with them. The Spaniards did examine and rifle of their contents the sepulchres in Cutifachiqui; and, upon the authority of the Ficlalgo of Elvas, ob tained from them three hundred and fifty pounds' weight of pearls--some of them formed after the simili tude of babies and birds. If the truth were known, or if an Indian had penned this account, we would be assured that De Soto and his companions, in their eager quest for treasures, without permission violated the graves and plundered the receptacles wherein were
1
PKAKLS AS OKNAMENTS.
46i)
garnered the most costly possessions of the natives. As a proof that the Indians did not willingly part with, these ornaments, but suffered the pillage through /ear of these strange and wanton men, we are informed that when the cacica, whom De Soto compelled to accom pany him with the intention of taking her to Guaxule ---the farthest limit of her territory--succeeded in mak ing her escape, she was careful to carry back with her a cane box filled with unboredpearls, the most precious of them all.
Luys Hernandez cle Biedina says that the gov ernor, while at this .town, opened a mosque in "which were interred the chief personages of that country: ".From it we took a quantity of pearls of the weight of as many as six arrobas and a half, or seven, though they -were injured from lying in the earth, and in tlie adipose substance of the dead." One of the saddest losses, ill the estimation of the relator, encountered by the expedition in the bloody affair at Manilla, was the destruction of the pearls which the Spaniards had been sedulously collecting during their wanderings in this strange land.
Fontaneda states that at the place "where Lucas Vasquez "went, seed-pearls were found in certain conchs ; and that between Havalachi and Olagale is a river the Indians call G-uasaca-esqui, which means in the Spanish language Rio de C'a-nas (river of canes). In this river, which is an arm of the sea, arid along the adjacent coast, pearls are procured from certain oysters and conchs. These are carried to all the provinces and vil lages of Florida, but principally to Tocobaja, the nearest town. The Indians of the town of A_balachi asserted that the Spaniards hung their cacique because he would not give them a string of large pearls which
470
ANTIQUITIES OF THE SOUTIIKBN INDIANS.
lie wore around his neck--the middle pearl being as big as the egg of a turtle-dove. Ribault frequently al ludes to the presence of pearls in the possession of the natives of Florida, and on one occasion saw the good liest man of a company of Indians with a collar of gold and silver about his neck from which depended a pearl " as great as an acorn, at the least." '
Father Hennepin * assures us that the Indians along the banks of the Mississippi wore " bracelets and ear rings of iine pearls which they spoilt, having nothing to bore them with but fire." lie adds: " They made us to understand that they have them in exchange for their calumets from some nations inhabiting the coast of the great lake to the southward, which I take to be the Oulpli, of -Florida." A member of the expedition of Sir Walter Raleigh collected from the natives of Vir ginia five thousand pearls, " of which number he .chose so many as made a fayre chaine, "which for their likenesse, and vniformitie in roundnesse, orientnesse and pidenesse of many excellent colours, with ec[ualitie in greatnesse, were verie fayre and rare." ' In the plates illus trative of the " Admiranda Narratio " and the " Brevis Narratio" the natives both of Virginia and Florida are represented in the possession of numerous strings of pearls of large size; and in his description of the " treasure or riches " of the Virginia Indians, Beverly says: " They likewise have some Pearl amongst them, and. formerly had many more, but where they got them is uncertain, except they found 'em in the Oyster Banks which are frequent in this Country." 4
1 " The Whole and True Diseovcrye of Terra Florida." Prynted at London by Rowland Hall for Thomas Ilackett, 15G3.
a "Neiv Discovery," etc., p. 177. London, 1(598. 3 " A Briefe and True Report of the New-found Land of Virginia," etc., p. 11. Fraucoforti ad Mremim. He Bry, anno 1590. 4 "History and Present State of Virginia," book iii., p 59. London, 1705.
MASNEK OF OBTAINING PEAKLS.
471
Wilson asserts that lie saw pearls " bigger than Rouncival Pease and perfectly round," taken from oysters on the Carolina coast.1
By far the most minute and interesting account of the manner in which the Indians obtained pearls and converted them into beads, is that furnished by Grarcilasso de ]a Vega. As this observation was made in the town of Ichiaha, which was in all likelihood located at or near the confluence of the Etowah and Oostenaula Rivers, and perhaps upon the very spot now occupied by the village of Rome in Georgia, the narra tive becomes all the more attractive:
" On the following day the Cacique visited the General," and gave him a string of pearls, two fathoms long. This present might have been considered valu able if the pearls had not been pierced, for they -were all of equal size, and as large as hazel-nuts. Soto ac knowledged this favor by presenting the Indian with some pieces of velvet and cloth, which were highly ap preciated by him. He then made inquiry of him with regard to fishing for these pearls, upon which the Indian replied that this was done in his province : that a great many pearls were stored in the temple of the city of Ichiaha, where his ancestors were buried, and that he might take as many of them as he pleased. The General expressed liis obligations, but observed that he would remove nothing from the temple, and that he had accepted his present only to please him. He desired to learn, however, in what manner the pearls were extracted from the shells. The Cacique replied he would send out people to fish for pearls all night, and that the foliowing day at eight o'clock " (jsia)
1 " An Account of the Province of Carolina," etc., p. 12. London, 1G82. 2 De Soto.
472
ANTIQUITIES OF THE SOUTHEBH INDIANS.
" Ills wish should be gratified. He at once ordered four' boats to be dispatched for pearl-fishing, with instruc tions that they should be back in the morning. In the mean time much wood was burned on the bank, producing a large quantity of glowing coals. AVheii the canoes returned, the shells were placed on the hot coals, and they opened in consequence of the heat. In the very first, ten or twelve pearls, of the size of a pea, were found and handed to the Cacique and the Gen eral, who were both present. They found them very fine, although the fire had partially deprived them of their lustre. "When the General had satisfied his curi osity, he retired to take his dinner. While thus en gaged a soldier came in who told him that in eating some of the oysters " (V^V) a caught by the Indians, a pearl had got bet-ween his teeth, which pearl being very fine and brilliant, he begged him to accept as a present for the Governess of Cuba.2 Soto very civilly declined the present, but assured the soldier that he was just as much obliged to him as if he had accepted his gift; and that he would endeavor to reward him some day for his kindness and for the regard he was exhibiting for his wife. He further advised him to keep his (intended) present and to buy horses with it at Havana. The Spaniards, who were with the Gen eral at that moment, examined the soldier's pearl, and some, who professed to be con/naisseurs- of jewelry, thought it was worth four hundred ducats. It had lost nothing of its lustre, as fire had not been employed in obtaining it." *
Garcilasso de la Vega, " Conquete de la rlori<]c,'A trad, par Eicholct
ANECDOTE Ol? JUAIST TEKRON.
473
During tlie course of the weary march, of the expe. dition through the mountains of Upper Georgia, the following circumstance is related by the same historian as having occurred:
" A foot-soldier, calling to a horseman who was his friend, drew forth from his wallet a linen bag in which were six pounds of pearls probably filched from one of the Indian sepulchres. These he offered as a gift to his comrade, being heartily tired of carrying them on his back, though he had a pair of broad shoulders ca pable of bearing the burden of a nrule. The horse man refused to accept so thoughtless an offer. ' Keep them yourself,' said he, ' you have most need of them. The Governor intends shortly to send messengers to Havana: you can forward, these presents and have them sold, and three or fonr horses and mares pur chased for you with the proceeds, so that you need no longer go on foot.' Juan Terron was piqued at hav ing his offer refused. ' Well,' said, he, ' if you will iiot have them, I swear I will not carry them, and they shall remain here.' So saying, he untied the bag, and whirling around as if he were sowing seed, scattered the pearls in all directions among the thickets and herbage. Then putting up the bag in his wallet, as if it was more valuable than the pearls, lie marched on, leaving his comrade and the other by-standers as tonished at his folly. The soldiers made a hasty search for the scattered pearls and recovered thirty of them. When they beheld their great size and beauty --none of them being bored, or discolored--they la mented that so many of them had been lost: for the whole would have sold in Spain for more than six
474
ANTIQUITIES OF THE SOTJTHERM" INDIANS.
thousand ducats. This egregious folly gave rise to a common proverb in tlie army that t there are no pearls for Juan Terrou.' The poor fellow himself became an object of constant jest and ridicule, until at last, made sensible of his absurd conduct, he implored them never to banter him further on the subject." '
It is the opinion of Colonel Pickett that the oyster alluded to by Grarcilasso "was identical with the mussel so common in all the rivers of Alabama. " Heaps of muscle-shells," says he, " are now to be seen on our river-banks where the Indians used to live. They were much used by the ancient Indians for some pur pose, and old warriors have informed me that their ancestors used the shells to temper the clay with which they made their vessels. But, as thousands of the shells lie banked up--some deep in the ground-- we may also suppose that the Indians in De Soto's time, everywhere in Alabama, obtained pearls from them. There can be no doubt about the quantity of pearls found in this State and Georgia in 1540, but they were of a coarser and less valuable kind than the Spaniards supposed. The Indians used to per forate them with a heated copper spindle, and string them around their necks and arms like beads." 3
Strange to say, Cabeca de Vaca makes no specific allusion to pearls, save that he was informed by the natives that on tlie coast of the jSoutTi Sea there *were pearls and great ricJies.
At the time of the Spanish invasion the pearl, as an ornament, was held in high esteem by the Mexican peoples ; and, upon occasions of state, its beauties were
PEARLS AS ORNAMENTS
475
invoked to enhance the magnificence of the apparel and lend additional lustre to the pomp of royalty. When Montezuma alighted from his regal palanquin, " "blazing with, burnished gold " and -overshadowed by a " canopy of gaudy feather-work powdered "with jewels and fringed with silver," to grant personal audience to Cortez, his ample cloak and golden-soled sandals -were sprinkled with pearls and precious stones.
Morales collected large "booty of gold and pearls from the Indians dwelling on the other side of the isth mus. The vanquished Cacique of Isla Rica brought as a peace-offering a basket curiously wrought and filled -with pearls of great beauty. Among them were two of extraordinary size and value. One "weighed twenty-five carats. The other was as " big as a musca dine pear, of Oriental color and lustre, and "weighed upward of three drachms."
The natives of Paria 1 possessed such quantities of
1 "Before the Spanish Conquest this was a smiling, happy coast, vexed occasion ally by Caribs, but otherwise a bright spot on the earth, where men, without mak ing much pretence to any thing that is elevated in human nature, lived peaceably and pleasantly enough, under the shade of their own cocoa-trees, looking out upon some of the grandest aspects of Nature. If they thought at all about the matter, they must have been delighted with the rich supplies of food which they obtained so easily from their oyster-beds. B(it the diseases of a creature apparently occu pying a low place in the scale of creation, were fated to be the means of dissolv ing the whole of Indian society in these parts, and of reducing large districts from a state of cultivation into a state of Nature, so that it is only conjectured now by the skilful naturalist, founding his conjecture upon the prevalence of some particular flower, that they were once cultivated.
" It is strange that this little glistening bead, the pearl, should have been the cause of so much movement in the world as it has been. There must be some thing essentially beautiful in it, however, for it has been dear to the eyes both of civilized and of uncivilized people. The dark-haired Roman lady, in the palmiest days of Home, cognizant of ail the beautiful productions in the world, valued the pearl as highly as ever did the simple Indian woman ; and a love for these glis tening beads came upon the Spaniards from two quarters--from the Romans who had colonized them, and from the Moors they had conquered. So general, indeed, was the love for pearls that it was to be expected that whatever country in the wide circuit of the whole world was cursed with an abundance of pearl-producing
476
ANTIQtTITIES OF THE SOUTHERN INDIANS.
fine pearls that the most sanguine anticipations were awakened in the breast of Columbus. Remembering the assertion of Pliny that pearls are generated from drops of dew which fall into the mouths of oysters, lie deemed no place so propitious as this coast for their growth, and multiplication. "When nearing the island of Ciibagua this admiral, as Charlevoix tells us, be held a number of Indians fishing for pearls, who, at the approach of the strangers, at once made for the land. A boat being sent to communicate with them, one of the sailors noticed many strings of pearls around the neck of a female. Having a plate of "Valencia^ware-- a kind of porcelain painted and varnished with gaudy colors--he broke it and presented the pieces to the Indian woman, who gave him in exchange a consider able number of her pearls. These he carried to the admiral, who immediately sent persons on shore well provided with Valencian plates and hawk's-bells, for which, in a little time, he procured about three pounds' -weight of pearls--some of which were of very large size, and were sent by him, afterward, to the sover eigns as specimens."
To Vasco Nuflez, Ttimaco gave jewels of gold, arid two hundred pearls 2 of great size and bsaiity, although
nil ; and it i r the si driven the jewellers of Europe wild with vexation) 011 the smooth brown arm or rich brown neck of his beloved."----(" The Spanish Conquest in America^ vol. it, p. 89. London, 1855.)
York, 1849. * Arthur Helps says : " Two hundred arid forty large pearls were presented on
this occasion." He continues: "The Spaniards could hardly contain their joy. O;io thing alone occurred to damp it. The Indians, not knowing better, were ac-
PEAKL--DIVERS.
477
they were somewhat discolored in consequence of the fact that.the oysters from which they were taken had been opened by fire. Observing the value -which the Spaniards set upon these pearls, the cacique sent a number of his men to fish for them. Certain of the Indians were trained from their youth to this purpose, so as to become expert divers and acquire the power of remaining a long time beneath the water. The largest pearls were generally found in the deepest water, sometimes in three and four fathoms, and were sought only iri calm weather. The smaller pearls were taken at the depth, of two and three feet, and the oysters containing them were often driven in quanti ties on the beach during violent storms. The party of pearl-divers, sent by the cacique, consisted of thirty Indians, with whom Vasco Nunez sent six Spaniards as eye-witnesses. The sea was so furious afc that stormy season that the divers dare not venture into the deep water. Such a number of the shell-fish, how ever, had been driven on shore, that they collected enough to yield pearls to the value of twelve marks of gold. They were small, but exceedingly beautiful, being newly taken and uninjured by fire. IMany. of these shell-fish and their pearls were selected to be sent to Spain as specimens.1
Oviedo commemorates the circumstance that this cacique, Tuniaco, subsequently furnished Vasco Nuilez -with a canoe of state, formed from the trunk of au enormous tree and managed by a great number of In-
custometl to open oysters by means of fire : this injured the color of the pearl ; and, accordingly, the Spaniards diligently taught the Indians the art of opening oysters -without fire, with far more diligence, indeed, than they expended in teach ing their new friends any point of Christian doctrine."--(" The Spanish Congest in. America," vol. i., p. 366. London, 1853.)
1 Irving's "Life and Voyages of Columbus and his Companions," vol. iii., p. 181. New York, 1849.
473
ANTIQUITIES OF THE SOUTHERN INDIANS.
dians. The handles of the paddles were inlaid -with small pearls--a fact which Vasco !Nunez caused liis companions to testify before, the notary that it might be reported to the sovereigns as a proof of the wealth of this newly-discovered sea.
In another bay of the Pacific coast this bold navi gator saw groups of islands abounding with pearls-- many of them as large as a man's eye. Davyd Ingram, during the " LandTravels " of himself and others in the years 1568 and 1569, from the Rio de Minas in the Gulf of Mexico to Cape Breton in Acadia, made the following observation: " There is in some of those Countreys great abundunce of Pearle, for in every Cot tage he founde Pearle, in some howse a quarte, in some a pottell, in some a pecke, more or lesse, where he did see some as great as an Acorn, and Eichard Browne, one of his Companyons, founde one of these great Pearles in one of their Canoes, or Boates, wch Pearle he gaue to Mouns' Champaine, whoe tote them aboarde his Shippe, and brought them to !Xewhaven in ffrunce." *
Without multiplying these references, we think sufficient historical evidence has been adduced to sat isfy the mind of the candid inquirer, and that beyond all reasonable doubt, that pearls were in general use among the Southern Indians ; that the choicest of them were the prized ornaments of the prominent per sonages of the tribes ; that the fluviatile mussels of various streams -were constantly and extensively col lected and opened for the purpose of procuring these gems, which, when obtained, were often pierced by means of heated copper spindles ; that the marine shells of the Atlantic, the Gulf of Mexico, and of the
1 " Documents connected with the History of South Carolina," edited by Plowden Charles Jennett Weston, p. 8. London, 1856.
PEABLS.
479
Pacific, yielded generous and beautiful tribute to tlie labor, skill, and taste of numerous and well-trained pearl-divers; and that these gems were found not only in the possession of the living, but also in large quan tities in the graves of chieftains and the sepulchres of the undistinguished dead. We are assured, moreover, of the eagerness with which the Spaniards sought after and preserved these treasures; and more than once do we hear expressions of disappointment at the discoloration and deterioration of the pearls caused by the action of fire, and their having been pierced. A present of pearls from the caciques to the conquerors was an earnest token of consideration, and the most acceptable pledge of friendship. It may be that the accounts which have reached us from the pens of the historians of these various expeditions and voyages, are somewhat extravagant with regard to the quantity and size of the pearls seen in the possession of the natives. It does not appear that many gems of this sort from Florida, Georgia, and Alabama, ever glad dened the eyes and enriched the coffers of the,home authorities, or graced the fair necks of Spanish beau ties. Most of them were observed and left amid the wilds of the Land of Flowers, "where the spring of per petual youth still conceals its life-giving waters be neath the shades of an untrodden forest. They were found and lost in that mythical region at whose upper end rose the fabled mountain from whose side flowed a stream of molten gold. And yet, in view of all the recorded observations, and in the light of subsequent investigations, we are not inclined to sympathize with those who regard -with equal incredulity the story of the Abalachi pearl, and the tale told by Sinbad the sailor of the vast treasures he saw in the valley of diamonds.
480
ANTIQUITIES OF THE SOUTHERN INDIANS.
AVith all clue allowance for the scope and effect of imagination, and a tendency to exaggeration highly developed in minds naturally alive to the marvellous and eager, in this terra -incognita, to perpetuate im pressions, which, when recounted at home, would ex cite the cupidity and awaken the intense interest of a people already familiar with the riches of Peru and Mexico and anxious to extend the hand of conquest over other regions in this ]STew A^orld, there is in the narratives of the career 'of De Soto, and in kindred re lations, ample proof that pearls of large size and of considerable value were in the possession of the Southern Indians dtiring the sixteenth century ; that their attention had been generally directed to collect ing xnargatiferous shells ; that by the simple process of heating them upon a bed of live coals they extracted the pearls from them ; and that they understood the art of piercing them with heated, copper spindles so that they might be strung and worn as ornaments around the neck, wrists, and ankles.
By the narrators of these primal recorded inter views between Europeans and the red-men we are informed that the Indians obtained their supplies of pearls both from marine shells and from fresh-water mussels. Some of the oysters on the Georgia and Florida coast are margatiferous. Many of them eontain seed-pearls. On sundry occasions specimens have passed under the writer's observation which were symmetrical in shape, as large as pepper-corns, and not wanting' in beauty. Some were quite big enough to have been perforated in the rude fashion practised by the Indians. They were, however, of a milky color and opaque. Neither in size nor quality did they answer the description of those spoken of in the Span-
PEAKiLS FRO5I MATURE SHELLS.
481
ish narratives. "We know that the Indians -who in habited the coast-regions of Carolina, Georgia, Florida, Alabama, and the more southern States, subsisted to a large extent upon oysters, clams, and conchs. This fact is to this day attested by the numerous and extensive kitchen-refuse piles and shell-heaps which abound upon the islands, along the headlands and upon the banks of salt-water creeks, and by the quantities of marine shells which were used as coverings for many of the sepulchral tumuli. These are .not the abraded driftshells cast upon the coast by the action of the waves, but are the perfect, uninjured shells from which the live animals had been artificially removed. Possess ing that passion for ornament so characteristic of all barbarous tribes, it excites no surprise that the Indians should, as they opened these marine shells, have care fully watched for pearls, and that from out the vast numbers consumed, year by year, C|uite a store of such gems should have been accumulated. But, if the shores of Carolina, Georgia, and Florida may not have afforded specimens of the larger and more highly prized pearls, we have only to look a little nearer the equator, and we will find pearl-bearing localities whose treasures fully gratified the taste of the savage and excited the cupidity of the civilized. Pearls eould have been here procured -which, in size and beauty, would corroborate the statements of the early navigators and justify, at least to a large extent, the seemingly extravagant representations of the strings of these gems encircling the necks, wrists, and ankles iu the oldest representa tions we have of the Southern Indians. In support of this opinion we have but to instance the trade in pearls which sprung up at an early period with the
482
ANTIQUITIES OF THE SOUTHERN- INDIANS.
islet of Cubagua, and at various points iu the Gulf of Mexico.
Suet were the trade relations existing "between the various tribes on this continent, so extensive their inter change of commodities, so general the office of runner or primitive merchantman, and so adventurous, in their larger canoes, the dwellers along the coast-regions of the South, it is not at all improbable that pearls from the islands and lower portions of the Gulf of Mexico and even from the Pacific Ocean may have found their way into the heart of Georgia and Florida and into more northern localities, to be there bartered away for skins and other articles, -which, in their turn, would subserve the purposes of this rude exchange of values. If, in the same ancient stone grave in !Nacoochee Val ley, we find a cassis from the Gulf of Mexico, a copper axe from the shores of Lake Superior, and stone imple ments the material for the manufacture of which -was necessarily obtained at no inconsiderable remove from this locality; if in the study of American archaeology we encounter, on every hand, proofs of an extensive and varied interchange of articles for use and orna ment, and the concentration in the ownership of a single individual of utensils and implements brought from places hundreds of miles apart, we surely do not overstep the bounds of probability when we suggest that the most admirable pearls among the Southern Indians once living within the present geographical limits of the United States -were obtained from marine shells native to the Gulf of Mexico. The replies of the Indians to inquiries addressed to them on this subject by Henuepin and others, and the presence in remote localities of beads, ornaments, and drinkingcups--all made of marine shells and conchs to this day
PEAKLS FROM FLUVIATILE SHELLS.
483
peculiar to the Gulf of Mexico--confirm the truthful ness of the suggestion.
But we are not confined to marine shells as the only or perhaps the 'chief source whence the Southern Indians derived most of their pearls. In all likelihood the fluviatile mussels contributed more freely than any other shells to the gratification of the ornamentloving masses. As we ascend the Southern rivers we observe, at various prominent points, relic-beds com posed in great degree of the fresh-water shells native to the streams. It is hardly an exaggeration to assert that no prominent stream is entirely devoid of them. The inland lakes of Florida afford similar evidences of the former occupancy of their shores by the aborigines, and even some ponds in Middle Georgia and .Alabama exhibit along their banks unmistakable signs of an cient refuse-piles into whose composition lacustrine shells enter largely.
As an illustration of the frequency and extent of these relic-beds along the banks of the rivers, we may instance those on the right bank o'f the Savannah Hiver, above the city of Augusta. Only one need be specifically mentioned, and this -will be found in Colum bia County, near the confluence of Great Kiokee Creek and the Savannah Kiver. Here, opposite a succession, of rapids in the river--a locality which would have afforded marked facilities for successful fishing in the manner adopted, by the Indians of this region--upon a bold bluff is an accumulation of fresh-water shells covering the surface of the ground to a depth varying from two to four feet, and extending nearly one hun dred yards in length, and more than a quarter of that distance in width. Intermingled with them may still be found the bones of large fishes, deer, turkeys, rac-
484
AMTIQTJITIES OF THE SOUTHERN HTDIAITS.
coons, bears, bison, turtles, squirrels, rabbits, and other animals and "birds, and also fragments of pottery, ar row and spear points, soapstone net-sinkers, crushingstones, axes, chisels, rude mortars and other imple ments, and various ornaments of clay and soap-stone. Here, then, was one of the favorite camping-grounds of the Indians. Hither they resorted for centuries, feeding upon fish, mussels, and game. This is but one of many extensive refuse-heaps of a similar charac ter which have attracted the notice of the "writer along the l)anks of the fresh-water rivers not only in Georgia, but also in Florida, Carolina, Alabama, and Tennessee. In these relic-beds no two parts of the same shell are, as a general rule, found in juxtaposition. The hinge is broken, and the valves of the shell, after having Taeen artificially torn asunder, seem to have l)een carelessly cast aside and allowed to accumulate at the very doors of the lodges, where, mixed with the debris of the encampment, iu the course of time they became heaped up to Buch an extent as to form these large shell-banks. In these early days the Southern rivers must have abounded with mussels. Their shells were sometimes used (as were the oyster, the couch, and the clam along the coast) in the construction of burial-mounds. Take, for example, that large tumulus located on Stalling's Island, in the Savannah River, a few miles above _A.ugusta, a description of which has already been presented. The river unios enter largely into its composition. The clay of which the Indians made their pottery was not infrequently mixed with par ticles of shells powdered for that purpose. It is also true that at least some of their shell ornaments were fashioned from the larger varieties of fluviatile shells found in the neighborhood. Evidently, therefore, the
PEARL--BEAKING SHELLS.
485
collection of fresh-water mussels must have occupied no little of the time and labor of the natives. That they subsisted largely, at certain seasons, upon them, as an article of food, admits of no doubt. Not a few of the unios of the Southern rivers, lakes, and swamps, are margatiferous. From the physical proofs enumer ated--aside from all historical testimony---where such quantities of shells were collected and opened, we may well believe that many pearls must have been found, and we incline the more readily to give credence to the statements of the Fidalgo of Elvas and the narrative of Garcilasso de la Vega. If it be true--as some have supposed--that the town of Cutifachiqui was located on the Savannah River, not veiy many miles below the site at present occupied by the city of Augusta, and if De Soto was standing on the bank of the Etowah when the Cacique of Ichiaha Ivindly sent his men to gather the mussels, and showed him how pearls were extracted from them, we still have, in the shell-heaps extant upon the banks of these streams, physical proofs of these ancient pearl-fisheries and ocular demonstrations of the verity of those relations.
With a view to ascertaining the precise varieties of shells from which the Southern Indians obtained .their pearls, the writer invited ail expression of opinion from several gentlemen of intelligence whose scientific pursuits rendered them familiar with the conchology of the United States. The following extracts from some of the replies which were received, will \>e found interesting, as throwing light upon the inquiry:
Dr. "William Stimpson, of the Chicago Academy of Sciences, expresses the opinion that the statements of the early Spanish historians with regard to the size of the pearls (as large as filberts) are incorrect. He says :
4S6
AXTIQtTITIES OF TUB SOUTIIEEK INDIANS.
" The pearls of the aviculcv--our only margatiferous marine genus--are very small, and those of the oyster, valueless. The Indians must have obtained their pearls from the fresh-water bivalves (unio and anodon}, wliicli abound in the rivers of Georgia, etc. These are usually small, but, in very rare instances, examples have oceurred reaching in diameter one-third of an inch."
" Most of the fresh-water mussels," writes Professor Joseph ]Lo Conte, " contain small pearls now and then. By far the best and largest number I nave seen -were taken from the Anodon Gibbosa (Lea), a large and beautiful sliell abundant in the swamps of Liberty County, Georgia--at least in Bull-town and Alataniaha Swamps. Some of the pearls taken from this species are as large as swan-shot. Of the salt-water shells I know not if any produce pearls except the oyster {Ostrea Virginiana). Pearls of small size are sometimes found in them." Professor ^William S. Jones, of the University of Georgia, says lie lias seen small pearls in many of the unios in Southern Georgia. I am informed "by Professor Wyman that, after a careful and extensive series of excavations in the shell-heaps of Florida, he lias failed to find in them a single pearl. "Tt is hardly probable," he remarks, " that the Spaniards could have been mistaken as to the fact of the ornaments of the Indians being pearls, but in view of their frequent exaggerations, I am almost compelled to the belief that there was some mistake; and, possibly, they may not have distinguished between the pearls and the shell beads, some of which would correspond with the size and shape of the pearls mentioned by the Spaniards."
Professor Joseph Jones, whose recent investigations have thrown much valuable light upon the contents of
PEAKL--BEARING- SHELLS.
4S7
the ancient tumuli of Tennessee, says: " I do not re member finding a genuine pearl in the many mounds whicli I opened in the valleys of the Tennessee, the Cumberland, the Harpeth, and elsewhere. Many of the pearls described by the Spaniards were probably little else than polished beads cut out of large sea-shells and from the thicker portions of fresh-water mussels, and prepared so as to resemble pearls. I have ex amined thousands of these, and they all present a laminated structure as if carved out of thick shells arid sea-conchs."
Mr. Charles M. "VVheatley is confident that there are " splendid pearls in Southern uuios." He instances the TJnio Blandingianus and the large old Unio jBuddifinus (Buckleyi) from Lakes George and Monroe in Florida, as pearl bearing. " In Georgia," he continues, " tlie large, thick shells of the Chattahoochee, such as the TInio Elliottii, would be the most likely to contain fine ones; but there is no positive rule, as an injured shell of any species will doubtless afford some: ir regular in most cases and of no value, but in some in stances worth from fifty to one hundred dollars." He mentions that he has received from the Tennessee Kiver, iu Alabama, fine round peai'ls both white and rose-colored.
From the response of Mr. John G. Anthony I ex tract the following : " I cannot so well answer your query as to what shells in Georgia and Florida are pearl-bearers, having never collected in the latter State and but little in Georgia, but I can say about Ohio what I presume will hold good in other States, that tie unios of various species furnish them tolerably abun dantly there. They are not confined to any one par ticular species, but are generally found in the thicker
488
ANTIQUITIES OF THE SOUTHERN INDIANS.
baashreneednlflomsotuoonrfsdetoeniphnoehnacadlavtnehearyolssuf,mospraolsltnhhdeoelsnl,sea,asn,ntidhemsopapuellgcahoicanelselyavcwescnuhociuchthnhtesposetefhceiimesnstnataegosr
to(nhKuaetnafJtioVuwfleiaastmthqeeiur.cem)a,naIaUnlrtn;eliceaoonloGdlefrcaaotlcnmtiealoikssspit(neBegcvaimeorrnveyeensro)ot,lwdfweshUnhicnteyhlilosItmhfFoearrulaelngodhinalieidnss
more or fewer pearls in it. Unio Torsu-s Unio Orbioulatus (Hildreth), and Unio finesque), Unto Undulatus (Bames), them in Ohio. I have seen about half a
(Rafinesque), C'ostatus (Raalso produce pint of beau
mbtBpmga-- lhmhapItmfiaeteneSaaamonieaeaycTaacnfykkrabcvoattourleepemeoyeteoerlu,sledurisyUluntansafrsjrophini,ulornetnireulaoaiTnidifwsytananLnsbowhitegcres-dhtdtcoeltrwerhotaCosoiehmhauscpnet,fto,eruetheieohhoerellsr,brqeeiorleefwiprtmblhswtpuabghlaemeutaonliatsuhihSaaatlnTvhaunbueapiscrleoaOeracalysloleeidcuiehtsrtrnshagitr(hltpgcnoreiSyhotIfioffelooahhoconseiracnnvganhfeu,akrsriyipeordeeadqnaetst,)barrrrnpsvu,,omecnruc-- ieaepeeewaadtmanotraeenemmstnualaardhrkacatslodcllshtesrieyeti-tneenlt.ncaoeinbe,iseerUnhdrtnndor.sirtnaea,indhnnL.npIaevncSogioeaohgrcoduIiyhfipouronuargireGetaeugndlvigrodhphiae.vsthegueefnie-axbghieosno,.stcaemrcoi"eheirrttvnteaaNzsrarigetivesaeseonhaocsswiosen,eetanrouneIrnesiwolsiyadhtmsaothcuhdhlp(oohoeprlafimneC,rallnuimru;ltliyocmrowlousohtbpsshkeetsetncsete,ruaoofwudadherrakwbutbbaoacahyahriellodonbenmelbnradoeyyrerliferd)tgyue.
Dr. Brinton observed along the Tennessee Eiver
many artificial shell-heaps and its tributaries. The
PEAEL--BEAKING- SHELLS.
489
Tennessee mussel ( TJnio Viryiniamisy is margatiferous, " and there is 110 doubt," says the Doctor, " but that it was from this species that the early tribes obtained the hoards of pearls which the historians of De Soto's exploration estimated by bushels, and which were so much prized as ornaments." 1
Dr. Kidder has recently pointed out the source whence at least small pearls and perhaps some fine specimens could have been obtained by the Indians of Florida, and in considerable quantities. ^11 the ^lnionidce of some of the fresh-water lakes of that State he has of late found not less than three thousand pearls --most of them small, but many large enough to be perforated and worn as beads. From one unio he took eighty-four seed pearls ; from another fifty, from a third twenty, and from several ten or twelve each. His examinations have hitherto been chiefly confined to Lake Grrifnn and its vicinity. He proposes soon, however, *o open the shells of Lake Okeechobee, which are larger, and there hopes to find pearls of su perior size and quality. It is said, but with what trnth cannot now be definitely affirmed, that upon one of the islands in this lake are the remains of an old pearl-fishery.
In view of the general use of the pearl as an orna ment by the Southern Indians, and of the quantities of lacustrine and fluviatile shells opened by them in various localities whither they resorted for the purpose of fishing and feeding upon these mussels, it seems singular that the pearl is not more frequently met with in the relic-beds and sepulchral tumuli of this region. ^W^e would expect to find them also in the refuse-piles, shell-heaps, and mounds of the coast. After an exami
ne Smithsonian "Report for 186(5, p. 357.
490
ANTIQUITIES OF THE SOTITIIEEN INDIANS.
nation of several fresh-water shell-heaps on the bants of the Savannah, and of others of a similar character in Alabama, Florida, and South Carolina, and after exploring many shell and earth mounds, particularly on the Georgia coast, the writer has failed, except in a few instances, to find pearls. These were obtained chiefly in an extensive relic-bed on the Savannah Kiver, abont twenty miles above Augusta, the largest being four-tenths of an inch in diameter, and all of them blackened by fire. It is, perhaps, not to be wondered at that many of the smaller earth-mounds on. the Greorgia coast do not contain pearls, because at the period of their construction the custom of burning the dead appears to have obtained very generally. So intense in some cases were the ilres tlieii kindled, that even hard stone axes and arrow-points were splintered. Under these circumstances it may be that the pearls were either immediately consumed or so seriously injxired as soon to crumble out of sight. Excluding this class of tumuli from present consideration, and cred iting the statements of the Eidalgo of Elvas and of others touching the large quantities of pearls found in Indian graves in the sixteenth century, we have been somewhat surprised that their presence has not been more frequently detected in relic-beds and tumuli in this region, in which there is no lack of shell-beads and other ornaments made of the same material. This ap parent absence of pearls tends in some measure to con firm the notion of those who entertain the belief that by the imaginative Spaniards many beads and orna ments made of the thicker portions of marine and fluviatile shells--carved, perforated, and brilliant -with their primal coloring--"were rated as pearls. The au thorities, however, are so numerous and direct, and
T-EABIS FOUND IN GBAVE--MOTJITDS.
491
the recent examinations into the contents of these tumuli and relic-beds have been so partial, that for one we cannot acquiesce, except to a qualified extent, in this opinion. Our impression is, that future and more minute investigations will reveal the existence of pearls, in various localities where the pearl-bearing mussels were collected, and where general inhumations occurred. Pei-forated pearls have been found in an ancient burial-ground, located near the bank of the Ogeechee River, in Bryan County, Georgia; and I am informed by the Reverend F. R. Goulding, that some twenty-five years ago, just after a heavy freshet in the Oconee River which had laid bare many Indian graves in the neighborhood of the large mounds on Povillain's plantation, he gathered on the spot fully a hundred pearls, of considerable size, some pierced, and others nnbored.
From the " altar " or " sacrificial" mounds, Messrs. Squier and Davis took a large number of pearl beads. By exposure to the heat, they had lost their brilliancy and consequent value as ornaments. Most of them were so much injured that they crumbled under the touch. The following is the account given of them in the " Ancient Monuments of the Mississippi Valley : " * " The peculiarities of their form, and their concentric lamellse, joined to the lingering lustre which some re tain, place their character beyond dispute. Several hundreds in number, and not far from a quart in quan tity, are in our possession, which retain their structure sufficiently well to be strung and handled. The largest of these measures two and a half inches in circumfer ence, or upward of three-fourths of an inch in diameter. They are of all intermediate sizes, down to one-fourth
1
1 Tagcs 232 and 233, vol. i., " Smithsonian Contributions to Knowledge."
1
492 ANTIQUITIES OF THE SOTJTJTEKX INDIANS.
of an inch in diameter. Most are irregular in form, or pear-shaped ; yet there are many perfectly round. They have been obtained from separate localities, sev eral miles apart, and from five distinct groups of mounds. Great numbers "were so much calcined, that it was found impossible to recover them, and a large number crumbled in pieces after removal from the mounds. It is no exaggeration to say that a number of quarts of pearls were originally deposited in the mounds referred to ; probably nearly two quarts were contained in a single mound."
AVithout expressing a decided opinion as to the precise locality whence these pearls were derived, it was evidently the impression of Messrs. Squier and Davis that for them a Southern origin should be sought.
From this examination it niay, we think, be fairly concluded:
first. That the possession by the Southern Indians of pearls, bored and uubored, at the time of primal intercourse between the white and red races, is clearly proven.
Second. That the use by the Indians of such orna ments was a matter not of recent, but of long standing.
TJiird. That evidence of the collection and employ ment of these gems was furnished not only by the ownership of living Indians, but also by the large and frequent accumtilations found in the graves and tumuli of the dead.
Fourth. That near the Gulf of Mexico and upon the Pacific coast lived trained divers whose occupation consisted in fishing for pearls.
Fiftti. That, in view of the trade-relations existing between the various American tribes, it is not at all
PEARLS.
493
unlikely that the finer specimens of pearls -worn as ornaments by tlie Indians of Florida, Georgia, Ala bama, Carolina, Louisiana, and more northern localities, were obtained from the islands and shores of the Gulf of Mexico, and perhaps even from the Pacific coast.
Sixth. That the fiuviatile shells and lacustrine unios of the Southern fresh-water risers and lakes were extensively gathered and opened by the natives both for the purposes of food and with a view of obtaining the pearls which they produced; and that from this source the Indians probably secured their principal supply of common pearls.
Seventh. That pearls from both marine and fresh water shells were greatly prized as ornaments by the aborigines, whose custom it -was to perforate them-- usually by means of heated copper spindles--and wear them, on strings around the neck, wrists, waist, thighs, and ankles.
HigJtih. That these gems were of such quality as to excite the cupidity of the early voyagers, and attract the marked attention of the various expeditions.
Ninth. That the marine shells of the Gulf of Mex ico and of some portions of the Atlantic and Pacific coasts, as well as the unios of the Southern rivers and lakes, could have supplied all the pearls represented by the early narratives as having been found upon the persons and in the temples and tumuli of the na tives.
Tenth. That the Spanish accounts of the quantity and size of the pearls seen in possession of the Indians during the fifteenth and sixteenth centuries, while they may be somewhat exaggerated, are not, in the main, to be regarded as unworthy of belief.
Eleventh. That the various shell-heaps along the
494
ANTIQUITIES OF THE SOUTIIEBN INDIANS.
coast and upon the banks of Southern streams, as well as the large quantities of shells, both marine and fluviatile, employed in the construction of sepulchral tumu li, should be reckoned as proofs of the general truthful ness of those narratives, and as furnishing indications of the local sources whence large numbers of pearls were probably derived.
And, lastly, that, in all likelihood, a careful exam ination of these shell heaps and mounds will, even at this day, disclose the presence of pearls.
CHAPTER XXII.
Primitive Uses of Shells.--Shell-Money.--Shell Ornaments.--Personal Decorations. --Concluding Observations.
tlie many relics which, escaping tlie disin tegrating influences of time and inherent decay, "bear present testimony to the fact that in former times they answered various artificial uses and "were freely ex changed in traffic among the Southern Indians, few are more "widely distributed then those made of shell. Copper from the prehistoric mines of Lake Superior, galena from beyond the Mississippi, mica from distant hills, silver and gold in small quantities, and numer ous worked flints and stones, are found in localities to which they should be utter strangers and in which their presence would never be expected but for the extensive interchange of articles which obtained among these primitive peoples. To the coast tribes the sea was the great treasure-house whence were derived abundant supplies with which they might constantly carry on a trade -with interior nations, and from them secure coveted products of tie mountains, chipped, rubbed, or beaten into well-known and desired forms of use and ornament. In the preceding chapter -we commented at some length upon the employment of
496
ANTIQUITIES OF THE SOUTHERN INDIANS.
pearls as gems for personal adornment and as articles . possessing tlie Highest commercial value among the redmen of the South. We have seen how diligently they were collected, how carefully they were perforated with heated copper spindles so that they could be worn as beads, and how extensively these beautiful offerings not only of the fresh-water mussels, but also of the shells of the Gulf of Mexico and the Southern seas, were distributed among tribes remote from locali ties whence they -were derived. These ornaments might very properly be considered in the present con nection ; but, in view of what has already been said on this subject, any further notice is here preterinitted. "We have also observed, upon an examination of the frequent and large refuse-piles, that the coast Indians and those dwelling near rivers and lakes, relied upon oysters, mussels, clams, and conchs, as important arti cles of food.
Although the labors of the primitive workers in shell were chiefly expended upon the manufacture of a convenient and well-recognized medium of exchange, and the preparation of various ornaments, in the do mestic economy of the natives sundry were the offices shells "were made to perform. Some of these we -will briefly enumerate:
I. They were employed as GOUGES, CHISELS, SCBAPEKS, and KNIVES.
In that rude period when men--almost entirely ig norant of the use of metals---were compelled from such objects as Nature placed within their reach to select those materials which -would most conveniently supply their mechanical requirements, the ancient artificers, avoiding the protracted labor necessary for the conver sion of stone fragments into implements of serviceable
SHELL SCRAPERS.
497
shape, found in the strong shells of the ocean and in many nuviatile mussels convenient tools, well formed, edged, and ready to hand.
In plate xii. of the u .A dmiranda N'arratio,'1 an Indian is represented with a conch busily engaged in scraping away the charred portions of the interior of a canoe which is being hollowed out by fire. From the part of the canoe upon which he is working the fire has evidently just been removed by his assistant, who, with a fan in one hand and a stick in the other, is kindling a name in another portion of the troughshaped boat. The" explanatory note informs us that by means of shells the bark was removed from the trunk destined for the canoe (" tune cortice eonchis qiii, busclam adernpto") ? and that; after it had been hol\lowed out by fire, its interior, with the aid of like im plements, "was scraped and rendered smooth ("restincto igne cochis scabunt, <fe nouo suscitato igne denuo aclurunt, atque ita deinceps pergunt, siibinde urentes & scabentes donee cyml>a necessariuna alueurn nacta sit.") *
The wooden spades and mattocks used by the Flor ida Indians in the cultivation of the soil -were made " with certain stones, oyster-shells, and mussels, where with also they made their bows and small lances, and cut and polish all sorts of wood that they employ about their buildings and necessary use." 3
Beverly a asserts that before the English supplied the Virginia Indians with metallic tools, their knives
1 " Admiranda Narratio," etc., Francoforti ad Moenum. De Bry, anno 1590. a " The Whole and True Discoverye of Terra Florida," etc., " written in French by Captain Ribaiilde, the first that wholly discovered the same, and now newly set forth in the English the xxx. of May, 1-663. Prynted at London by Rowland Hall, for Thomas Hackett." 3 " History and Present State of Virginia," book ill., chap, siii.j p. 60. London, 170!?.
498
ANTIQUITIES OF THE SOUTHERN INDIANS.
consisted either of sharpened reeds or shells, and that with these and sharp stone-axes " bound to the end of a stick and glued in with turpentine," they formed l>ows of locust-wood, and cut and notched their arrows.
The oyster-shell was employed as a* scraper in dressing hides.'
Many of the clam, oyster and mussel shells of the Southern waters were well adapted to the uses of scrapers and gouges; and the supply of such natural tools was at all times accessible, and limitless in quan tity. So common -were they, that near the coast they were not regarded of value sufficient to w^arrant their inhumation with the dead.
II. As DKINKIWG-CUPS.--The use of certain conchs as drinking-cups seems to have been general among the Southern Indians. When the Floriclians, in the sixteenth century, would deliberate upon grave affairs, the chief men were "wont to assemble in the public place, where, upon a semicircular wooden bench, they all took their seats. The king or inico appeared also, and occupied his place in the centre, where was a seat elevated above the rest. A-t his command certain women prepared the casino-. Upon a given signal from the cacique, the cup-bearer offered this hot decoc tion in a capacious shell first to the king, and then to the noted personages who were present, each drinking in the order of his rank."
In plate xix. of the " Brevis Narratio," widows, in token of their grief, are strewing their hair upon the graves of their dead husbands. Upon each grave are
1 "Natural History of North Carolina," etc., Brickell, p. 365. Dublin, 1737. Lawson's "History of Carolina," etc., pp. 338, 339. Raleigh reprint, 1860.
2 " Turn poeillator primum Regi hoc dococtum calidum in capace concha praibct, dcinde (sic impcranteRege), omnibus alijs ex ordine, in illaipsa concha." "Brevis Narratio," etc., plate xxix. Francoforti ad Medium. De Bry, anno 1591.
SHEIi DEIKKIXG--CUPS.
499
seen the bow, quiver, spear, and shell clrinking-cup of the deceased.1 XJpon the demise of a king1 or priest, for three days did the members of his tribe gather around his tomb and mourn and fast. About the base of the tumulus numerous arrows were stuck: in. the ground, while upon its top was placed the shell from which, he "was accustomed to drink.2
In many of the burial-mounds of Georgia conchs are found which "were doubtless used as drinking-cups, and placed there at the period of the inhumation in obedience to that -well-established custom .-which sur rounded the dead with articles of value, ornament, and convenience, that there should be no lack of them in the spirit-land. From, some of them the axes have been entirely removed. In the stone graves of Nacoochee Valley more than one cassia flammed was seen. In each instance the interior whorls and columellas had been carefully cut away, so that these large tmivalves formed capacious and serviceable vessels.
Similar relics were observed by Professor Joseph Jones in the stone graves of Tennessee, and they have been found in ancient tumuli in several of the South ern States. Sometimes these shells were, at great pains, divided longitudinally. In the neighborhood of the coast the FyTula, perversct seems to have been the common drinking-cup, and, in its natural shape, handily supplemented the calabashes and fictile -ware in minis tering to the simple wants of these primitive peoples. JVIr. Haywoocl 3 says that at the annual feast of Har vest the Southern Indians sent to those of their
1 " Maritorum arma, conchas ex qutbus bibcbant." 2 " Brevis Narratio," plate xl. "Defuncto aliquo Rege ejus Provincial, inagna solemnitate sepclitur & ejus tumulo crater, e quo bibere solcbat, iinponitur, defixis circa ipsiim turuulum raaltis aagittis," 3 "Xatural and Aboriginal History of Tennessee," p. 156. Nasbvilie, 1S23
500
ANTIQUITIES OF THE SOUTHERN INDIANS.
number who were sick and unable to participate in tlie solemnities and festivities of the occasion, old con secrated shells, full of the sanctified, bitter casina.
III. As SPOONS.--Clam and cockle shells were ex tensively used in this way. Generally the half-shell, in its natural state, sufficed; but, in many instances, a handle, just wide enough to be conveniently grasped by the thumb and forefinger, was cut in the side near the hinge. In this way hot food might be scooped up without bringing the fingers in contact with it.
Lying upon the mat by the side of the woman, one of these shell spoons is figured in plate xvi. of the "Admirancla !Narratio." ' Such shells also served a good turn in scaling fishes. In the writer's collection are fine specimens taken from the grave-mounds of Tennessee.
IV. As AGKICTJLTURAL IMPLEMENTS." V. As RATTLES.--These were made of the shells of the land-tortoise," or of conchs from -which the interior whorls and columellas had been removed and pebbles, beans, or beads placed in them. By means of deer-skin thongs they were fastened to the outside of the legs. In dancing, every saltatory movement was accompa nied by a corresponding jiugle, and thus each motion called forth a certain sort of rude music. VI. As RECEPTACLES OB SHKINES EOK IDOLS.--Dr. Troost had in his collection a large casais flammed whose interior whorls and columella had been, entirely removed, and the front of the shell opened so as to permit the entrance and enshrining of a small image
1 See also Bevcrly's " History and Present State of Virginia," book iii., chap, iv., p. 17. London, 1705.
a Loskicl's "North American Indians," pp. 66, 67. London, 1794. 3 Adair'd " History of the American Indians," pp. 169, 170. Lonjon, 1775.
SIIEIjL--MONEY.
SOI
in a kneeling posture. That idol was within the shell when it was ploughed up, and is figured in aittt. on page 361 of volume i. of the "Transactions of the American Ethnological Society."' This may be an exceptional case, "but it is well authenticated and wor thy of specific mention in this connection.
VII. As AIT El/EinSXT OF STRENGTH AND DURABILITY
IN THE JVLi-iNXrFACTURE O-P EARTHEN-WARE.----Kor this
purpose shells were reduced, by pounding, to a fine powder and mixed with the clay. The mass, moist ened with water, was then, carefully kneaded and subsequently formed into the desired vessel. As we. have, however, in the chapter devoted to an examina tion of the pottery of the Southern Indians alluded to this use of shells, we refrain, from further comment.
VIII. As MONEY.--Ignorant of the relative worth of metals, and, in the manufacture of serviceable and ornamental articles, treating gold, silver and copper simply as malleable stones, it was necessary that the Indians in the interchange of various. commodities should agree upon something which by common con sent should be .regarded and accepted as the represent ative of fixed values. Accordingly, they selected w~hat is now generally known as wampum, or shell-money. The term wampum is said to be an Algonkin word, signifying zvhitG--such being the prevailing color of the beads. The ordinary wampum beads * are cylin drical in shape, varying from the sixth to a quarter of an inch in length and being about the eighth of art inch in diameter. They are of two varieties, the one
1 New York, 1845.
the Mission of the United lireUiren," etc., p. 26. London, 1794. 3 These beads are variously known as wampiimpeage, wampeage, pease, wam
pum peak, peak, scawun, sea want, ronoak, etc., etc.
502
AKTIQTJITIE3 OF THE SOTTTHEKN INDIANS.
wliite, and tlie other blue of purplish-black--the latter being the more valuable (see Figs. 1 and 2, Plate XXX.).
On the Virginia coast, as we are informed, in the " Westover Papers," the species of conch-shell is found of which the Indian peak is made : " The extremities of these shells are blue, the rest being white, so that peals of both these colours are drilled out of the same shell, serving the natives both for ornament and money, and are esteemed by them beyond gold and silver."
Beveriy' thus describes what he quaintly terms the treasure or riches of the Virginia Indians: " The Indians had nothing which they reckoned Riches before the English went among them, except Peak, Roen-oks, and sueh-like trifles made out of the C-unk Shell. These past with them instead of Gold and Sil ver, and serv'd them both for Money and Ornament. It was the English alone that taught them first to put a value on their Skins and Furs, and to make a Trade of them.
" Peals is of two sorts, or rather of two colours, for both are luade of one Shell, tho of different parts ; one is a dark Purple Cylinder, and the other a white; they are both made iu size and figure alike, and com monly much resembling the English Huglas, but not so transparent nor so brittle. They are wrought as smooth as Glass, being one-third of an inch long, and about a quarter, diameter, strung by a hole drill'd thro the center. The dark colour is the dearest, and distiuguish'd by the name of Wampom Peak. The English men that are call'd Indian Traders value the
1 " History and Present State of Virginia," book iii., chapter xii., p. 58. Lon don, 1705.
C3 ^?
tt
F
f
.9-
1
S3
QS U'*:
/: )EB51 ^
SHEIi--MOJTE3T.
503
Wampom Peak at eighteen pence per Yard, and the "white Peak at nine pence. The Indians also mate Pipes of this, two or three inches long, and thicker than ordinary, which are much more valuable. They also mate Huntees of the same Shell, and grind them as smooth as Pealc. These are either large, like an Oval Bead, and drill'd the length of the Oval, or else they are circular and flat, almost an inch over, and one third of an inch thick, and drill'cl edgeways. Of this Shell they also make round Tablets of about four inches diameter, which they polish as smooth as the other, and sometimes they etch or grave thereon Cir cles, Stars, a Half-Moon, or any other figure suitable to their fancy. These they wear instead of Medals before or behind their N"eck, and use the Peak, Huntees, and Pipes for Coronets, Bracelets, Belts, or long Strings, hanging down before the Breast, or else they lace their Garments with them, and adorn their Tomalicvwlcs and every other thing that they valne.
" They have also another sort which is as current among them, but of far less value; and this is made of the Cockle shell, broke into small bits with rough edges, drill'd through in the same manner as Beads, and this they call lioenolee, and use it as the Peals.
" These sorts of Money have their rates set upon them as unalterable, and current as the values of our Money are.
"The Indians have likewise some Pearl amongst them, and formerly had many more, but where they got them is uncertain, except they found 'em in the Oyster Banks, which are frequent in this Country."
The money of the Carolina Indians, says Lawson,1 " is of different sorts, but all made of shells which are
1 " History of Carolina," etc., p. 315. Kaleigii reprint, I860.
504
ANTIQUITIES Olf THE SOUTHERN INDIANS.
found oil the coast of Carolina, which are very large and hard so that they are very difficult to cut. Some English smiths have tried to drill this sort of shellmoiiey, and thereby thought to get an advantage; but it proved so hard that nothing could be gained. They oftentimes make of this shell a sort of gorge, which they wear about their neck iu a string; so it hangs on their collar, whereon sometimes is engraven a cross or some odd sort of figure which comes next in their fancy. There are other sorts valued at a doe skin, yet the gorges will sometimes sell for three or four buck skins ready dressed. There be others, that eight of them go readily for a doe skin ; but the general and current species of all the Indians in Carolina, and, I believe, all over the continent as far as the Bay of Mexico, is that which we call Peak and Ronoak ; but Peak more especially. This is that which at New York they call wampum, and have used it as current money amongst the inhabitants for a great many years. This is what many writers call poroelan, and is made in New York in great quantities, and with us in some measure. Five cubits of this purchase a dressed doe skin, and seven or eight purchase a dressed buck skin. .An Englishman could not afford to make so much of this -wampum for five or ten times the value; for it is made out of a vast great .shell, of which that country affords plenty; w7here it is ground smaller than the small end of a tobacco pipe, or a large wheat straw. Four or five of these make an inch, and every one is to be drilled through, and made as smooth as glass, and so strung as beads are, and a cubit of the Indian measure contains as much in length as -will reach from the elbow to the end of the little finger. They never stand to question whether it is a tall man or a short
SHELL--MONEY.
OOO
man that measures it; but if this wampum peak be black or purple, as some part of that shell is, then it is twice the value. This the Indians grind on stones and other things till they make it current, but the drilling is the most difficult to the Englishmen, which, the Indians manage with a nail stuck in a cane or reed. Thus they roll it continually oil their thighs with their right hand, holding the bit of shell with their left ; so, in time, they drill a hole quite through it, which is a very tedious work; but especially in making their 1'onoak, four of which wTill scarce make one length of wampum. The Indians are a people that never value their time, so that they can afford to make them, and never need to fear the English will take the trade out of their hands. This is the money with which you may buy skins, furs, slaves, or any thing the Indians have; it being the mammon (as our money is to us) that entices and persuades them to do any thing, and part with every thing they possess, except their chil dren for slaves. As for their wives, they are often sold, and their daughters violated for it. V/ith this they buy off murders; and whatsoever a man can do that is ill, this wampum will quit him of, and make him, in their opinion, good and virtuous, though never so black before." *
Alluding to the passion of the Southern Indians for ornaments, Adair' remarks : " Before we supplied them with our European beads, they had great quan tities of wampiim (the Buccinum of the ancients), made out of conch-shell by rubbing them on hard stones, and so they form them according to their liking.
1 Compare Dr. BrickelPs " Natural History of North Carolina," p. 337, ft seq. Dublin, 1787.
2 "History of the American Indiana," etc., p. 170. London, 17G5.
506
ANTIQUITIES OF THE SOUTHERN" INDIANS.
" With these they bought and sold at a stated cur rent rate, without the least variation for circumstances either of time or place; and now they will hear noth ing patiently of loss or gain, or allow us to heighten the price of our goods, be our reasons ever so strong, or though the exigencies and changes of time may require it. Formerly four deer-skins was the price of a large conch-shell bead, about the length and thickness of a man's fore-finger; which they fixed to the crown of their head as an high ornament--so greatly they valued them. Their beads bear a very near resem blance to ivory."
"When Cabeca cle Vaca set out upon his trading ex pedition he carried with him from the Gulf coast " coiies and other pieces of sea-snail, conches used for cutting," and " sea-beads." These he traded away to the Indians inhabiting the interior, and in exchange received from them and brought back with him " skins, ochre with which they rub and color the face, hard canes of which to make arrows, sinews, cement and flint for the heads, and tassels of the hair of deer that by dyeing they make red." Wherever he journeyed, while thus emploj'ed, he received fair treatment at the hands of the natives, who--to use his own language--" gave me to eat out of regard to my commodities. The inhabitants were pleased when they saw me, and I had brought them what they wanted." On various occasions shell-beads were offered as presents by the Southern Indians to the Spaniards. 1 In this way they sought to propitiate their powerful invaders, and the gift was, in their esti mation, among the most valuable of all their posses sions. Among the articles regarded as " great riches "
1 " Relation of Alvar Nuiiez Cabeca dc Vaca," translated by Buckingham Smith, pp. S3, 86, 145, 148, ISO, 194. Now York, 1871.
SHELL--MOTfET.
507
by the inhabitants of Pacaha, Biedma enumerates
" beads made of sea-snails." * This shell-money was also extensively manufactured
by some of the Northern Indians, and for a consider able time circulated freely in the New-England colo nies, in New York, Pennsylvania, and elsewhere. Sev eral interesting accounts of the value and use of this currency in that region, during the early clays of Euro pean colonization in America, have been preserved.
The New-England Indians, writes Roger Williams,'' " are ignorant of Europe's Coyne ; yet they have given a name to ours, and call it JHoneash from the English Money. Their owne is of two sorts; one white, which they make of the stem or stocke of the Per-iwi/ncle, which they call Meteauhok, when all the shell is broken off: and of this sort six of their small Beads (which they make "with holes to string the bracelets) are cur rant with the Englisli for a peiiy. The second is black, incling to blew, which is made of the shell of a fish which some English call liens, Poquaukock, and of this sort three make an English perry. They that live upon the Sea-side generally make of it, and as many make
as will. " The Indians bring clowne all their sorts of Furs
which they take in the Countrey, both to the Indians and to the English for this Indian Money: this Money the English, French and Dutch trade to the Indians, six hundred miles in severall parts (North and South from New England) for their Furres, and whatsoever they stand in need of from them, as Corne,Veuison, etc.
" This one fathom of this their stringed money,
anjo dc Soto," translated by Buckingham
OOS
ANTIQUITIES OF THE SOUTHERN INDIANS.
now worth of the English but five shillings (sometimes more) some few yeeres since was worth nine and some times ten shillings per Fathonie. . . . Their white they call Wompam (which signifies white) ; their black SucJcdiiboclc (^Sacki signifying blacke). . . . Before ever they had Avjle-blades from Europe they made shift to bore this their shell money with stone, and so fell their trees \vith stone set in a "wooden staff, and used woden bo^ves,1'!
The money of the Massachusetts Indians is de scribed by the Kev. Cotton Mather as consisting of " little beads with holes in them to sti'ing them upon a bracelet, whereof some are white, and of these there go six for a penny. Some are black or blue, and of these go three for a penny. This wampuni, as they call it, is made of the shell-fish which lies upon the seacoast continually." Nathaniel Morton * intimates that the Plymouth colony first acquired a distinct knowl edge of the value and profit of the trade in loampampeag from the Dutch in 1627, and denounces the " base ness of sundry unworthy persons " who, in exchange for this shell-money, furnished the Indians with "guns, powder, and shot." So firm a hold, however, did this wampum--as a standard of values and as a convenient medium of exchange--soon take upon the commercial mind of the New-Englanders, that at an early period it was, by special enactment, treated as currency and made a legal tender in payment of debts not exceed ing specified amounts. The wampum-trade was also farmed out to a company which, for the privilege of the monopoly, obligated itself to pay into the colonial treasury of Massachusetts one-twentieth of all that was secured.
1 "K"eiv England's Momoriull," etc., p. 67. Cambridge, 1669.
SHELL--MONEY.
509
In bis " Account of two Voyages to New Eng land " Josselyn asserts that the natives made wam pum so cunningly " that neither Jew nor devil" could counterfeit it. Subsequently, hoAvever, as Mr. Stevens * properly remarks, this proved to be an idle boast, for a spurious imitation, very closely resembling real wampum, was introduced by the fur-traders at so low a price that the whole Indian country was soon flooded with it, destroying at once the value and meaning" of real wampum.
Buriiaby,3 who made his observations in 1759 and 1760, describes the current money among the Indians as " made of the clam-shell consisting within of two colours, purple and -white, and in form not unlike a thick oyster-shell. .The process of manufacture is very simple. It is first clipped to a proper size, which is that of a small oblong parallelopiped, then drilled, and afterwards ground to a round, smooth surface, and pol ished. .The purple wampum is much more valuable than the white--a very small part of the shell being of that colour." a
Without multiplying authorities, it may be safely asserted that this shell-money was manufactured along the Atlantic coast from Maine to Florida, and on the Gulf coast certainly as far south as Central America. The use of this circulating medium was undoubtedly very general among the agricultural tribes east of the Mississippi River. The ancient sepulchral tumuli of Georgia, Tennessee, Florida, and of other Southern States, as well as those located in the valley of the
1 "Flint Chips," etc., p. 458. London, 18TO. 3 " Travels through the Middle Settlements in North America," etc., p. 60. London, 1775. 3 Compare Carver's " Travels," etc., p. 362. London, 1778. LoskieTs "His tory," etc., p. 26. London, 1794.
510
ANTIQUITIES OF THE SOUTHERN INDIANS.
Ohio and in valleys tributary Iboth to it and to the Mississippi from the east, when opened, fully corrobo rate the historical narrative, and afford physical proof that this product of the skill and the patience of the coast tribes--sought and obtained through, trade-rela tions--was thus, and by means of subsequent migra tions, widely disseminated among the red-men dwell ing far in the interior. After he crossed the Missis sippi, Mr. Catliii 1 saw but very little wampum among the prairie Indians. " Amongst the numerous tribes," he states, " who have formerly inhabited the Atlantic coast and that part of the country which now consti tutes the principal part of the United States, wampum has been invariably manufactured and highly valued as a circulating medium."
"West of the Rocky Mountains, however, some of the tribes " make use of various coloured shells, ground to ail oval or nearly round shape." Belts of wampum were also regarded as standards of value, and. accord ing to these standards they exchanged property among themselves and with the traders.2 Among the Indians of the Northwest coast the Dentalium formed.a cur rency. 3
Taking the place of money, and constituting an acknowledged medium of exchange, these wampumbeads served also as favorite and valuable decorations. Broad belts, variously and elaborately ornamented with such beads, were delivered at one time as title-deeds upon the alienation of a tract of land, at another time as solemn tokens in ratification of a treaty of peace;
1 " Illustrations of the Manners, Customs, and Condition of the North Ameri can Indians," vol. i., p. 223, note. London, 1848.
2 Hunter, " Memoirs of a Captivity," etc., p. 294. London, 1823. a J. K. Lord, "Naturalist in British Columbia," vol. ii., pp. 25, 26. Stevens' " Flint Cliips," p. 488, et seq. London, 1870.
SHELL ORNAMENTS.
511
again, as pledges of friendship, as sacred attestations of an tittered vow, and as records of memorable events. In the latter case, each string of beads possessed an historical significance and was as intelligible as the knotted cord of the guipu.
IX. As ORNAMENTS.--While the shape and char acteristic peculiarities of what is commonly called the wampum bead are readily recognized and clearly de fined, it seems probable, at least among the Southern Indians, that all the various forms of shell beads, pen dants, and ornaments, were highly prized both for per sonal decoration and as objects of barter. Rarely have I seen the purple or black wampum within the limits of Georgia, -while hundreds of the white have been taken from sepulchral tumuli in various portions of the State. The Southern Indians, without doubt, expended no little time and toil in the manufacture of these shell ornaments. Consequently, the results of their taste and industry are numerous and interesting. Sharing in that passion for personal decoration which, in all ages, has so thoroughly possessed the breasts of both civilized and savage, they found in the pearly nacre and bright colors of marine and iluviatile shells the choicest material for the fabrication of beads, pen dants, gorgets, armlets, pins, and various ornaments with which to bedeck their persons and habits. To these ornaments a twofold value appertained--the one inherent in the intrinsic beauty and durability of the shells themselves, the other born of the skill, ingenuity, and labor involved in their manufacture. Strings * of these shining and carefully-polished beads adorned the ears, necks, shoulders, elbows, arms, knees, ankles,
1 " Admiranda Narratio," plates iii., vi. vii., viii., xvi,, xviii., xxi. u Brevls Narratio," plates viii., xiv., xvi., xxxiv, xxxvii., xxxviii., xxxix.
512
ANTIQUITIES OF THE SOUTlTtyRN" INDIANS.
wrists, waists, and robes of these primitive peoples, and were used to enhance the beauty, dignity, and riches of their idols. Both males and females delighted in the ownership of personal ornaments; and, when the grave "opened to receive those who claimed them, these prized possessions were deposited "with the dead, that, amid the well-watered fields of fairer huntinggrounds, the departed might not lack the companion ship of those things which pleased them most and con stituted their chief treasures here. Tumuli and oblit erated graves are now the storehouses whence are ob tained supplies of these ornaments. Upon most of them the lapse of years, fire, and the moisture of the earth have wrought sad changes, and they are often but crumbling, discolored mockeries of former symme try and beauty.
Dwelling under warm sMes, which permitted them to pass the greater part of the year in a state of almost entire nudity, the Southern Indians delighted in paint ing their bodies with the most brilliant colors they could command. Their persons being uncovered, the fullest opportunity -was afforded not only for the dis play of skin ornamentation in various lines and curious devices, but also for the exhibition upon any part of the body of necklaces, gorgets, and sundry articles of shell, bone, and stone jewelry, if indeed that word may be properly used to describe these representatives of barbaric fancy. Hence the taste for personal decora tion was more general and pronounced among them, than among their more northern brethren, whose prin cipal labor in this regard was bestowed upon the orna mentation of their clothing.
The prevailing varieties of the shell beads found within the limits of Georgia are represented in Figs. 14-19, Plate XXX.
SHELL BEADS.
513
"With tlie exception of tlie disk-shaped beads, all are perforated longitudinally, the diameters of the bores varying -with the size of the ornament--seldom, how ever, exceeding a quarter of an inch. Some of them are perforated both longitudinally and transversely. It is evident that, at the period of their manufacture, they were all carefully polished; and while many have, -with the lapse of years, been converted into a soft, white, chalky substance, others still retain their smooth surfaces, and in their present appearance closely resem ble ivory, for which substance they were sometimes mistaken by the early observers. The column and Avails of the Stroinbus gigas were freely used in the construction of the largest of these beads, not a few of which still bear the trace of the natural canal. Those of the elongated shape vary in length from a quarter of an inch to two inches and a half, and in diameter from one-sixth of an inch to one inch. The disk-shaped beads vary in thickness from the twelfth to the sixth of an inch, and in width from a quarter of an inch to an inch and a quarter. The forms varied with the fan cies of the manufacturers, some beads being round, others ovoidal, others tubular, and others still, disk-
shaped. Both Adair * and Lawsoii = unite in stating that the
natives manufactured these beads out of conch-shells, and formed them into the desired shapes by rubbing them on hard stones. Before the introduction of me tallic implements, Roger Williams * says the Indians " made shift to bore this their shell money with stone;"
1 "History of the American Indians," etc., p. 170. London, 1775. a "History of Carolina," p. 310. Raleigh reprint, I860. BrickcH's "Natural
History of North Carolina," p. 339. Dublin, 173*?. 3 "A Key into the Language of America," etc., p. 148. London, 1043.
514:
ANTIQUITIES OF THE SOUTHERN INDIANS.
and, during the progress of the journey of SurveyorGeneral Ijawson,' he observed the Carolina Indians drilling their beads by means of a nail stuck in the end of a cane or reed. The drill was rolled on the thigh with the right hand--the bit of shell being held in the left--and so, in the course of time, after the expendi ture of much patience, the perforation was accom plished. When we consider the amount of tedious labor necessarily involved in shaping, boring, and polishing these beads, we are prepared to appreciate the reason, why they came to be regarded by the na tives among their most precious treasures. It is not probable that the heated copper spindles, which the Spanish historians declare were used for the perfora tion of pearls, could have proved serviceable in punc turing these pieces of shell. The larger beads -were drilled from, opposite ends, the perforation being smaller in the centre than, at the inception of the bore. There is no reason why at least some of them should not have been drilled in the manner commonly adopted for boring stone. ICither a solid or a hollow wooden drill, aided by sharp sand and water, would have com passed the desired object; and in the case of the diskshaped, round, and ovoidal beads, a drill made of a triangularly-pointed flint flake would have answered every purpose.
Among the Southern Indians, upon the authority of Adair," in former times " a large conch-shell bead about the length and thickness of a man's fore-finger would purchase four deer skins." Beads of this sort were greatly valued, and were " fixed to the crown of the head as high ornaments."
1 "History of Carolina," p. S16. Kaleigh reprint, 1860. ' * "History of American Indians," etc., p. 170. London, 1777.
SHELL BEADS.----SHELL GOEGETS.
515
Du Pratz 1 describes the ear-rings of the Indian women of Louisiana as being " made of the center part of a large shell, called burgo, which is about the thickness of one's little finger; " and Father Hennepin,2 in his account of the customs of the natives of Louis iana and Mississippi, states that "Women and Men, but above all, Young Girls, wear Necklaces of Shells about their ISTecks, of different Figures. They have also a sort of Shells as long as one's Finger, and hol low like Pipes, which serve them for Pendants to hang
in their Ears." While the longer varieties served as pendants and
head-ornaments, the smaller were strung and worn as , necklaces, bracelets, anklets, armlets, or used as deco\rations for moccasins, belts, and their clothing gener*ally. The number of these beads found in a single
tumulus is surprising, and shows how many of them were at times owned by one individual. 1
In obedience to the taste and skill of the Southern Indians, the shell assumed ornamental shapes other than those represented l>y the beads. Prominent among them are the gorgets--two varieties of which are here represented (see Figs. 3 and 4, Plate XXX.). These, suspended by a string, -were -worn about the neck. Lawson * alludes to the existence of this class of ornaments among the Carolina Indians in his day, and comments upon the high commercial esteem in which they were held. He also calls attention to the fact that thereon was sometimes " engraven a cross, or some odd sort of figure -which comes next in their
1 " History of Louisiana," vol. ii., p. 232. London, 1763. 2 " Continuation of the New Discovery," etc., p. 80. London, 1698. 3 Compare Roger Williams' "Key into the Language of America," etc., p. 149.
* "History of Carolina," etc., p. 315. Raleigh reprint, 18GO.
516
ANTIQUITIES OF THE SOUTHERK INDIAKS.
fancy." The Sotitliem Indian priest wore upon his breast an ornament " made of a white conch-shell with two holes bored in the middle of it, through which he ran the ends of an otter-skin strap and fast ened to the extremity of eacli a buck-horn white but ton.' The natives of Virginia * manufactured round shell " Tablets of about four inches diameter," which they polished as smooth as their peak, and upon "which they etched or graved " Circles, Stars, a Half-Moon or any other figure suitable to their fancy." It is quite probable that the " figures of children and birds made of pearl," said by the Portuguese narrator to have been found by De Soto in the temple at Talomeco, were nothing more than beautiful gorgets, the personal orna ments of the departed caciques and chieftains of Cutifachiqui who were there interred.
The largest of these ornaments (Fig. 3, Plate XXX.), it will be perceived, is elliptical in shape--its diame ters, measured in the direction of the major and minor axes, being respectively four inches and three inches and a half. It is about the eighth of an inch in thick. ness. In the upper edge are two holes--rather more than half an inch apart--by means of "which it "was suspended. The open-work and ornamentation are, we think, to be regarded rather as the expressions of the rude fancy of the "workman, than as indications of any intelligent design or pictographic idea. These gor gets were, at the period of their manufacture, carefully polished, and the ornamentation occurs on the inner or concave surface. This, then, was the side intended for display. The interior of the shell being lined with an
1 Adair's " History of the American Indians," p. 84. London, 1775. 3 " History and Present State of Virginia," book iii., chapter xii., p. 59. Lon don, 1703.
SHELL GORGETS AND ATtMLETS.
517
iridescent nacre, and that surface having been by Na ture polished beyond all art, was far more beautiful than the exterior, and was consequently selected for exhibition. This we believe to be the true interpreta tion of the thought of these peoples in the use of such ornaments. Some of the gorgets are bored only in the centre ; others have holes both in the upper edge and in the central portion, which would indicate that they were sometimes suspended, and at other times worn as fixed ornaments attached to the head-dress or cloth ing. In form, size, and ornamentation, these relics do but express the individual fancy of those by whom they were made; and while in the accompanying illus tration we have indicated only two prevalent types, 1 to wit, the elliptical and circular, we might mention \ others which are square, ovoidal, stellate, parallelogrammic and irregular in shape, some with and some without scalloped edges, and others still which, care lessly constructed and with a single hole ill the centre, suggest the idea that they were designed as shell but tons. 1 Closely allied to the gorgets are the shell arm lets and anklets.
Such is the peculiar shape of these ornaments that they appear by nature adapted to the curvature of the arm or leg.
By means of a thong passing round the limb and through the holes, they could have been readily worn in any desired position. Many were probably used, at pleasure, either in the manner we have suggested or as gorgets suspended from the neck or ears.
Another variety of shell ornaments found in the
518
ASTTIQTJITIES OF THE SOTJTHEEJf INDIAN'S.
Southern mounds is that which may be designated by tlie general name of Pins. Two forms are liere rep resented, the one pointed at either extremity and tumescent in the central portion (see Fig. 5, Plate XXX.), the other with one end terminating in a large, wellformed head, and the other tapering to a blunt point (see Figs. 6 and 7, Plate XXX.). Those with heads were made from the columellas of some big univalve, such as the Strombus gigas. The extreme length of the pin numbered 6, in Plate XXX., is five inches and a half, one inch of that distance being occupied by the head, which is an inch and a quarter in diameter. The shank is an inch and a half in circumference; and, while tapering somewhat, is blunt at the point. Relics pre cisely similar in shape were fashioned of soapstone. From the same tumulus pins made both of shell and stone have been taken. The pointed pins are usually smaller, seldom exceeding three inches in length, while those -with heads vary in length from an inch and a half to six inches. These ornaments were at the time of their manxvfaeture highly polished in every part. While their precise use is open to conjecture, -we may safely conclude that they were intended as objects of display and personal decoration. 1
Shells -were frequently "worn as ornaments without any material alteration of their natural forms. A_mong the Southern Indians the oliva and the marginella (see Figs. 8 and 9, Plate XXX.) -were extensively used as necklaces, bracelets, and anklets. The apices of the former -were cut or rubbed off, aud the backs of the latter ground so as to make a second hole or perfora-
UNFINISHED SHEKL BEADS.
519
tion by means of which a thread of some sort could be conveniently introduced, and thus any desired number of the shells securely strung.' In several localities we have found the columns of large sea-shells cut off at the required lengths, partially fashioned and imperforate, which were evidently obtained in this imperfect condition from the primitive shell-merchantmen and kept for polish and completion, at some future time, by the purchasers. (See Figs. 10, 11, and 12, Plate XXX.) Cabega de "Vaca alludes to a trade in such articles, and the banks of the Ocmulgee near ]VIa.con, and of the Chattahoochee far up among the beautiful valleys of Cherokee Georgia, as well as the sites of many old Indian villages, bear present testimony to the truth of his narrative arid to the extensive character of this ancient traffic in unfinished shell beads.
We might enumerate other shell trinkets, but they are matters rather of curiosity than of archaeological value.
Beads were also manufactured of stone, clay, bone, and wood. Those of stone were generally made of soapstone, are globular in shape, and about three-quar ters of an inch in diameter. The clay beads are circular in form, the upper and lower sides being flat, are per forated through the centre, are from a quarter to half an inch in thickness, and vary in size from half an inch to an inch and a half in diameter. When these disklike beads were strung in quantities, only the edges
1 Compare " Transactions of the American Ethnological Society," vol. i., p. 360, New York, 1845. "Anelent Monuments of the Mississippi Valley," p. '233. Wil son's " Prehistoric Man," pp. 129,141, London, 1865. Stevens' "Flint Chips," p. 454, London, 1870. "Smifhsonian Report for 1868," p. 404. Venegas' " Na tural and Civil History of California," vol. i., pp. 71, 73, London, 1759. Heunepin's "Continuation of the New Discovery," p. 80, London, 1698. "Narratives of the Career of Hernando de Soto," etc., translated by Buckingham Smith, p. 252, New York, 1866.
O20
ANTIQUITIES OF THE SOUTIIKRN INDIANS.
would appear. Bone beads were cut in desired lengths fi-oni the wing-bone of a large bird or from the small bone of the leg of a deer or other animal. Their perfora tions are longitudinal, and the ornaments when finished were carefully polished. {See Fig. 13, Plate XXX.)
Human teeth, and the teeth and claws of bears and the spurs of the turkey-cock, were perforated and worn aa pendants. Youths frequently bedecked themselves with bracelets made of the ribs of deer, which were boiled, bent into the desired shapes, and then polished so as to look like ivory. 1
In comparatively recent sepultures European beads of glass and porcelain are not infrequent. Black, blue, white, and red, are the predominating colors. Many are enamelled, and are evidently Venetian in their origin. \Vith these European beads the white wam pum and other shell beads are often found inter mingled. .Portions of strings of rosary-beacis also occur, which were doubtless obtained at an early period through religious commerce with the Spanish priests.
Secondary interments upon the tops and sides of an cient tumuli, and many Indian graves in Gherokee Geor gia contain ornaments of silver and brass, consisting of corrugated bracelets, ear and finger rings, pendants, buckles, clasps, bosses, and gorgets. In most of them we recognize how sedulously the European manufacturer pandered to the barbaric tastes of these primitive peo ples.
In the Etowah Valley gold beads have been found which, were clearly the handiwork of the natives. Copper pendants also are occasionally unearthed in Nacoochee and other valleys in Upper Greoi'gia. In all instances of this character, as we have already re-
' Da PraU, "History of Louisiana," vol. ii., p. 233. London, 1753.
ORN AME^TTS.
521
marked, the metal was treated without the application of heat, and was simply hammered, into the desired shape. Among the Aboriginal tribes of this region, prior to commerce with Europeans, the use of metallic ornaments and implements was limited.
Adair 1 assures us that in the olden time quartzcrystals (or, to use his own ]anguage, " such coarse diamonds as their own hilly country produced ") were freely used. They were fastened by means of deersinews to the hair, nose, ears, moccasins, and various parts of the dress. The truth of this statement is attested l)y the contents of many mounds which we have opened. Mr. Atwater is correct when he sup poses that the circular aggregation of crystals which he figures on page 233 of Volume I. of the " Archseologia Americana," was worn as an ornament, AVe have in our collection a beautiful specimen, of this character taken from a Georgia burial-mound located two hundred, miles away from any point whence the quartz-crystals could ha^e been, obtained. The holes in the lobes of the ears of the women were small, but the men were in the habit of cutting out the entire interior of the lobes of their ears, and then inserting large tufts of buffalo's-wool mixed with bear's-grease, so as to distend the aperture to the utmost degree." Into these, when healed, they would introduce bunches of beautiful feathers, large rings, joints of cane gaudily colored, and the inflated bladd.ers of fishes. 3 In the ex planatory note accompanying plate xxxviii. of the "Brevis Karratio," we are informed that both men and -wom en wore these fish-bladders in their ears, and that when
522
ANTIQUITIES OF THE SOUTHERN" INDIAN'S.
inflated they stone like pearls. Sometimes they were colored red, and then they resembled carbuncles. The first joint of an eagle's leg1, 1 with the talons attached, formed a favorite ear-ornament with the Southern war rior.
Cabega cle Vaca 2 observed some of the Florida Indians.with their nipples and under lips bored, and wearing pieces of cane in the openings. Pendants of various sorts from the nose and under lip "were custom ary, and it may be that lip-stones, after the fashion of the Mexicans, were also used as personal decorations.
CONCLUSION-.
As it comports uot with the plan proposed and adopted in the execution of this work, we refrain from entering upon a discussion of the interesting inquiry, AVhence came the recl-meii who tirst peopled this por tion of North America ?
Our object has been to present a general descrip tion of the Southern Indians as they appeared when the Europeans first ventured among them, and to in terpret their relics in the light of early recorded obser vations and of customs not obsolete at the date of the Spanish, French, and English colonizations.
Comparing the manners and temper of the South ern Indians with those of the more Northern tribes which he visited, Father Hennepin pronounces the former " Civil, Easie, Tractable and capable of instruc tions," while the latter are declared " mere Brutes as fierce and cruel as any wild Beasts."" Enjoying those physical blessings which are bestowed by warm skies, luxuriant vegetation and abundant animal life,
COIfCXUDISG OBSERVATIONS.
623
the Southern Indians were in great measure relieved of those perpetual struggles for covering and food which have such a decided tendency to harden the condition of the savage, embitter his existence, and render him an Ishmaelite even among his own fellows. To the soft airs which surrounded them and the gener ous trees which alike in -winter and summer threw their protecting arms about them, to the food-treasures of the w^ater and the forests--ever present to supply with little effort their simple wants--and to the spon taneous productions of a fertile soil, were these peo ples largely indebted for the pleasure-loving disposi tion and the imaginative temperament they possessed, and for the gentle lives they were permitted to lead. Exempt from trials incident to a rigorous climate and an inhospitable country, they -were able to devote much of their time to amusements and social enjoyment, and to the development of a degree of taste and skill ill manufacture superior to that exhibited by their North ern neighbors. Upon a careful comparison of the an tiquities of the Southern nations with those of the Northern tribes, we think a greater variety and excel lence of manufacture, a more diversified expression of fancy in ornamentation, a more careful selection, of beautiful material, a superior delicacy and finish in the fabrication of implements, both chipped and polished, a more pronounced exhibition of combined labor in the erection of tumuli, a more despotic form of gov ernment, a greater permanency of seats, a more liberal expenditure of care and attention in the cultivation of the soil, a more decided system of worship, and a more dignified observance of significant festivals and funeral customs may fairly be claimed for the former. "We are acquainted with no region north and east of the liio
524
ANTIQUITIES OF THE SOOTHERS INDIANS.
Grrande iii which the earliest exhibitions of skill and taste in the manufacture of implements and. ornaments of stone, shell, and bone, are more varied and attrac tive, where pipe-making claimed such special attention, and where the antique pottery is indicative of such diversity of form and ornamentation, and possessed of such homogeneousness of composition and durability.
Our observations have been, perhaps, too general, and not sufficiently minute, to satisfy either our own wish or the intelligent desire of those in quest of specific information touching the interesting subjects of which we have essayed to treat. Sufficient has been said, however, \ve-trust, to afford the inquirer a tolerable conception of the antiquities of the region which has formed the field of research. So manifold are the exhibitions of fancy and use among the stone implements, so frequent the modifications of welldefined types, and. so varied the traces of early con structive skill, that were we to pursue the investiga tion with that detail which characterizes the recent and most valuable work of Mr. Evans upon the an cient stone implements of Great Britain, we would scarcely be able to assign a reasonable limit either to the descriptions or.to the illustrations which would be sug gested. A particular consideration of hammer-stones, mauls, sling-stones, whet stones, and. of minor relics, as well as of unfinished objects, such as nirit chips, wasters, etc., has been omitted. "Flakes and splinters of silicious stone, "whether flint, jasper, chert, iron-stone, quartzite, or obsidian, are to be fonnd in almost all known, countries, and belong to all ages. They are, in fact, the most catholic of all stone implements, and have been in use ' semper, ubique et ab omnibus.' " *
1 Evans' " Sto-.ic Implements, etc., of Great Britain," p. 257. London, 1S72.
BULLET--MOULD.
O^O
The objects wliicli we have selected for illustration are designed to convey a suitable idea of prevailing types, not abnormal forms. For this purpose we have used, such originals as are in our own possession, re fraining entirely from reproducing; a single illustration which has appeared elsewhere. For the genuineness of these relics and the accuracy of the drawings we stand personally pledged. Of the relics obtained through early commerce with Europeans, and found in the graves of modern Indians, much might be said: but these belong to a transition period and do not properly claim present attention. There are other relics, the product of the labor and the ingenuity of the reel-men in their effort to satisfy wants suggested by personal intercourse with these strangers. Of these a BTjiLET-MOCLD, taken from an Indian grave in Oostenaula Valley, may be mentioned as an example. Made of soapstone, it contains thirteen matrices for running shot and balls, varying in size from a swanshot to an ounce-ball. These cavities are carefully and regularly cat, and the entire arrangement is most creditable to the workman who, in the absence of a metallic bullet-niould, was able, in the exercise of his native ingenuity and skill, to manufacture an article out of a material with which he was entirely familiar, so cleverly answering the use which contact with the whites had taught him to understand and to require.
American archaeology is as yet in its infancy, and there are, on every hand, inviting fields in which in telligent observers may reap rich harvests. If these pages shall minister to the entertainment of the gen eral reader, and contribute aught of value to the in formation of the careful student, the pleasurable labors of the author will not have been entirely in vain.
GENERAL INDEX.
Aconitlms, mound at, 119. Adair, James, 8, 19, 8G, 115, 251,
273, 800,' 309, 833, 341, 418, 505. Adultery, punishment of, 66, 67. Adze, 277Agamemnon, 119. Agriculture, 396-320. Agricultural implements, 301-803,
500. Agricultural labors, 41, 30T. . Alexander the Great, 120. Alibamons, 31. Alyattes, 119. Amexias, 45. Amulets, 372, 373. Anasco, 247. " Anklets, 517. Appalatcy, 229. Archdale, Governor, 2. Archery, 245-250. Armlets, 517. Armories, at Talomeco, 20. Arrows, of the Florida Indians, IS;
manufacture of, 245-259. Arrow-heads, 240; general distri
bution of, 240-242 ; where manu factured, 242, 243 ; articles of commerce, 243 ; of what materials made, 244, 247, 249, 250, 256; typical forms of, 254, 265-267; size of, 257; how attached to shafts, 257; how manufactured, 259-265. Arrow-makers, 243. Arrow-shafts, stones for rounding, 366, 367. Arrow-stems, of cane, 255. Artachies, 119. Articles of dress, 61.
Ash, Thomas, 250, 322, 422 Assembly-room, 15. At water, Caleb, 436. Awls, 291, 292. Axes, stone, 269-286 ; general dis
tribution of, 269; how made and
perforated or ceremonial, 2S1284.
B.
IJaegert, Jacob, 363. Ball-play, 96-98. Bartrain, "William, 3, S, 20, 41, 123,
150, 178, 190, 216, 316, 328, 417. Baskets, manufacture of, 225. Beads, of European manufacture,
235-287, 520; of shell, 511-514; of stooe, bone, and clay, 519 ; of gold, etc., 520. Beckwith, Lieutenant, 263. Beverley, 250, 270, 362, 470, 502. Biedma, Luys Hernandez de, 235. 469. Black-drink, 11, 15. Blackmoor's teeth, 33T. Blankets, of buffalo's-wool and tur key-feathers, 87. Blow-guns, 256, 257. Bolzius, Rev. Martin, 417. Bone houses, 113, 191. Boos-ke-ta, feast of the. 99, SOS307. Borers, 291, 292. Borysthenes, Scythian tombs on, the "banks of the, 119. Bossu, Captain, 184, 256, 271, 323.
TERAL INDEX.
527
Bows, 245-257. Bread, preparation of maizo for
making, 310, 311. Breech clouts, 74, 75, 81, 86. Brickell, Dr. John, 329. BrintoQ, Dr. D. G., 230. Browne, Sir Thomas, 118. Brown's Mbimt, fortification on,
163-165. Bullet-mould, 525. Burial of the dead in a sitting pos
ture, 183-185. Burial-ground on the Georgia coast,
205-207. Burial-urns, 454^4-56.
Calabashes, 445, 464. Calumet", 386-393; typical forms
of, 404-408; ho\v drilled, 408, 409. Calumet-dance, 388-390. Canoe, ancient, dug up in Savannah River swamp, 53--57. Canoes, 53-61; manufacture of, by Virgiuia Indians, 55. Capital punishment, 13. Carpets, 86. Oassis riammea, 233. Catawbas, 2. Catlin, George, 261, 869, 462. Catlmite, 407. Celts, 278-281. Chaouanons, 222. Charlevoix, Father, 23, 428. Chateaubriand, Viscount do, 204. Chenco, game of, 346. Chcrokoes, 2, 7, 8 ; territory of the, 6; physical characteristics of the, 9. Chcrokoe chief, funeral obsequies of a, 115. Chiekasaws, 2, 3. Chieftain-mounds, 183-188. Chisca, 229. Chisels, 286-288, 496. Choctaws, 2, 256; origin, of the, 5. Chungke-garne, 96, 341-346, 356, 357. Chunky-yards, 178-181. Circular earthwork on the head waters of theOgeechee River, 148. Clavigero, 227. Columbus, 476. Conjurer, the office of, 28. Cooking, 308-311.
Copper, treated as a malleable stone, 47, 227, 231 ; use of; among tbe Southern Indians, 227-233 ; axes of, 281 ; implement of, from stone grave in Naooochee Valley, 225-227 ; from the E to wall Val ley, 232 ; pendants of, 233 ; rods of, 2-32.
Coreal, Francois, 29, 362. Costume, 71-89. Council-house, 11. Counting, 101. Co we, council-house of, 125. Craven, Governor, 3. Crawfish, mode of taking, 33G. Creeks, 2, G. Creek Confederacy, territory of the,
3-6; tribes composing the, G. Cremation, 101, 189-192, 411. Cupping-tubes, 361. Cutifachiqui, the Cacica of, 24, 71,
148, 149, 247.
D.
Dablon, Ffltbei-, 387. Daggers, 267. Dances, various kinds of, 92-96, 388
-390. Davis, Dr. E. II., 232. DeBrahm, William Gerar, G, 39, 83,
421. De Brv, 11, 209-211. De Soto, 24, 25, 142, 349, 235, 4G8. I>entalmiD, nse of tbe, 510. Devil, worship of the, 21. Discoidal stones, 341-358 ; not ex
empt from sepulture, 343, 3*6, 347; various forms of, 348-351 ; applied to secondary uses, 351, 352. Divorce, 66. Domenech, Abb6 Em, 351. Doi-antes, Andres, 229. Drift-implements, 293-295. Drilling in stone, 408. Drills, solid and hollow, 408-410. Drinking-cups, 233, 498. Drums, 90, 91. Dug-outs, or wooden boats, manu facture and use of, by the South ern Indians, 53-61. Duinont, 462. Du Pratz, M. Le Page, 78, 103, 211, 272, 302, 345. Dwellings of the Florida Indians ; 35, 37 ; of the Virginia Indian?,
52S
GENERAL INDEX.
36 ; of the Carolina tribes, 37,38; of the Georgia tribes, 39.
Dyeing, art of, 03, 88.
among the Chickasaws, 114; among the Cherokees, 114--116.
185; carefully observed, 116, 190! Funeral-scaffolds, 112.
Ear-ornament?, 88, 515, 521. Earth-walls, 212. Edistoes, 2. Efau-Haujo, 420. Elevations for chieftain-houses, 122,
126.
Elf-stones, 251. Elvas, the Gentleman of, 18, 25,
142, 229, 246, 300. Emetic, prepared from calcined
shells, 29. Enchanted mountain, 377. Enclosed \york on Plunkett Creek,
147. Etowah idol, 432-434. Etowah mouads, 136-143. European axes, 285. Evans, John, 524.
F.
Family or tribal mounds, 189. Feasts, monthly, amongtheNatchcz,
99, 100. Feather mantles, 01, 87. Festivals, 99, 100. Fire, veneration of, 21 ; new lighted
at the feast of the Boos-ke-tau, 99. Fishing-, various modes of, 325--340. Fish-gigs, 329, 330. Fish-hooks, 326, 327. Fishing-plummets, 328. Fish-preserves, 142, 143, 156, 175,
325. Fish-traps, 330-334. Flutes, 90. Fontaneda, 409. Food, animal, 42, 43; vegetable,
44, 308, 311. Fort James, ancient monuments
near, 123. Fortification on Brown's Mount,
163-165. Funeral customs, 101-117, 132, 183
-185, 190-192, 203; among the Choctaws, 104, 112, 113, 190; among the Natchez, 105 ; among the Virginia tribes, 108; among the Carolina tribes, 108-111, 184 ; among.the Muscogulges, 113, 184; among the Alibamons, 114, 184;
G.
Gallatin, Albert, 3, 6. Galphin, Fort, 151. Game, chungke, 96, 341-346, 356,
357 ; of the javelin, 354 ; of the pole, 345 ; of nettecawaw, 346 ; of the spear and ring, 355. Gaming, 98, 99. Gare.ilasso, 122, 471. Gardens, private, 42, 299-301. Georgia, original boundaries of the Colony of, 7. Gold beads, 47. Gorgets, 515-517. Gouges of stone and bone, 287, 288 ; of shell, 496. Government, system of, obtaining among the Southern Indian*, 10. G-ranaries, public, 41, 307. Grapes, 45. Graves, 113 ; veneration and attach ment for, 110, 117, 204, 205. Grave-mounds, 101-105. Grooved axes, 274-278. Guyachoya, Cacique of, 24-. Gygtean Lake, 119.
II.
Hammers, 265. Hand-axes, 278-281. Hand-nets, 335, 330. Hariot, Thomas, 17, 26, 30, 76, 310. Harvesting the maize, 307. Hatchets, 281-284. Hawking, Colonel Benjamin, 14, 65. Haywood, 84-86, 216, 346, 360, 437. Head-warriors, 16. Hennepin, Father, 230, 270, S28,
386, 470. Hephajstion, 120. Herodotus, 119. Hickory-nut-oil, 45, 316. High-priest, office and duties of
the, 19. Hired mourners, 111. Hitchittees, origin of the, 4. Hoes, 301. Homer, 120. Horn bells, 92. Hospitality of the Southern Indians,
42.
529
Hot-houses, 15. Human sacrifices, 23, 24. Hnrnaii remains found in a cave in
Tennessee, 84-86. Hunter, John !>., 401. Hunting, 322.
I.
Idol-pipes, 401-403. Idol-shrines, 481, 500. Images, 140, 146, 480-440. Image-worship, 22, 413-415, 417-
419, 423-430. Immortality of the soul, belief in
the, 21. Incised trenches on Stone Mountain,
880. Intaglios, 63, 377-399. Iron, no knowledge of, 47.
J.
Jaoiianas, 29, 30. Jars, 457. Javelin-game, 354. Jefferson, Thomas, 193, 436. Jones, Prof. Joseph, M. I)., 221-223,
233, 268, 280, 439, 458. Jugglers, 31.
Khvasa, the idol, 20, 420. Knives, flint, 290, 291, 4U6.
L.
Lafiteau, 271. La llontan, 192. Lake Superior, ancient mining on
the shores of, 232. Land, tenure of, 40. Laudonniere, 249. Lawson, Surveyor-General, 2, 31,
SO, 109, 328, 423-425, 504, 505. Leaf-shaped implements, 291, 3C2. Lee, Colonel Henry, 151. Lip-stones, 88, 522. Longfellow, H. "W., 384. Loskicl, 270, 332. Lnbhock, Sir John, 268, 414, 444. Lyon, Caleb, 263.
M.
Maize, cultivation of, 297-301 ; har vesting of the, 307 ; preparation S4
of, for food, 808-314; varieties of, 310. Maize-crashers, 314.
Marginella, use of the, as a head, 518. Marriage, 66-6$.
Matting, cane, 225. Manilla, 211. Maxwell, Major J. A., 166. Mechanical labor of the Southern
Indians, 46-53.
Medicine-men, 28-33. Medicine-tubes, 359-365. Medicinal plants, 34. Megalithic monument?, absence of,
Kesrier's Mound, 160-174. Mica membrauacea, 376. Mico, office and powers of the, 11--
13; duties of the Creek, 14; cabin of the, 15 ; selection of an assist ant for the, 14. Mining, ancient, in Duke's Creek Valley, 48 ; in Yallev-Eiver Val ley, 48.
Mirrors, 376, 377. Moats, ancient, 136, 146, 155, 170,
171.
Money, shell, 501-511. Mortars, 309-314. Moscoso, Lilys de, 24. Mound - builders, 135, 101, 176 ;
skull of one of the, 160. Mound-building, 118; in Georgia,
121 ; within the historic period, 130-132.
Mounds, on the Colonel's Island, 129 ; in the Etowah Valley, 136142 ; in the valley of Little Shoul der-bone Creek, 143 ; on Plankett Creek, 147; on Mason's plantation, 153-157; on the Ocnralgee River opposite Macon, 158-162; on Lamar's plantation, 162; on JSfcssier's plantation, 166 -174; of observation and retreat, 181,182; sepulchral, 183; chief tain or priest, 183; family or tribal, 189 -192 ; on the low grounds of the Rivanna, 193 ; on Stalling's Island, 197; of shell, 195-200; of stone, 202; in Nacoochee Valley, 213 ; at tne junction of the Etowah and Oostenaula Elvers, 253.
Muscogee Confederacy, 2. Museogccs, origin of the, 4; physi
cal characteristics of the, 9, 10. j
530
Music, 90. Musical instrun
K.
Nacoochee Valley, mound in, 213; stoue graves in, 214-224.
Narvaez, Panpliilo de, 116. Natchez. 2: the sun among the, 22--
31. Nets, 143, 326, 335-337, 339. Net-sinkers, 337-340. Nettecawaw, game of, 34G. New fire, origin of, 420. Nichols, Captain J. IT., 214. Nipples, bored, 88. Nose-ornaments, 88. Nuiiez, Vaseo, 476. Nut-stones, 315-320.
O.
Offering of the stag, 21, 22. Ogletborpe, General, 3,86, 131, 188,
189, 417, 421. Oliva, the use of the, as a head, 518. Open-air workshops, 242. Orestes, 119. Ornamental tubes, 365. Ornamentation of primitive pottery,
444. Ornament?, 71-89; of shell, 511-
510; of European manufacture, 520. Ortiz, Jnan, 116.
P.
Paeaha, 142. Painting, 63. Palanquin, use of the, 72, 73. Paria, the coast of, 475. Patroclus, the burial of, 120. Peace, how concluded, 15. Pearls, 149; use of, as ornaments,
among the Southern Indians, 4(!7-- 494; large numbers of, found in the possession of the natives and in the graves of their dead, 467-- 481 ; method of procuring, 471, 472, 476 ; diving for, 477 ,- trade in, 482 ; obtained from both ma rine and nnviatile shells, 481-490; found in relic-beds and ancient graves, 491. Pear-shaped stones, 371, 372. Perforated axes, 281-284. Jersonal pr the dead, 102, 185.
Pestles, 314. Pendants, of copper, 233; of stone,
370, 371. Physicians, 28-33. Pierced tablets, 367-370. Pins, of shell, 233, 234; of soap-
stone, 233, 234. Pipe-stem carrier, 409. Pipes, 383-412 ; origin and uses of,
383-385 ; how made, 408; how drilled, 407-410 ; calumets, 386393; typical forms of, 404-407; idol-pipes, 401--403 ; ordinary or common pipes, 394; typical forms of, 410-412. Pits, 187. Plates, stone, 373-376. Platters, 374. Pocahontas, dance of, 94, 95. Poisoning fish, 327, 333. Pol ishing-st ones, 292, 293. Polygamy. 68. Population, aboriginal, 128. Pots of terra-cotta, 457. Potter's wheel, use of the, unknown, 47. Pottery, manufacture of, 46 ; general description of, among the Southern. Indians, 441-466 ; an index of the degree of civilization, 440; histor ical value of, 443; ornamentation of, 444, 458, 459 ; use of, 445451; manufacture of, 451, 461464, 501; . kilns for baking, 452, 453 ; various sizes of articles of, 453; burial-urns, 454-456; pots, 456, 457; flat-bottomed jars, 457; typical forms of Southern, 458 ; vessels of soapstone, 460; de stroyed by cremation, 465, 466. Priapus, worship of the, 439. Priest-mounds, 187. Public buildings ia Creek villages, 15, 16. Public deliberations, 15. Public granaries, 41. Public overseer, 41.
Q.
E.
Battles, 91; of gourds, 91 ; of ter rapin-shells, 92, 500.
Ban, Prof. Charles, 64, 218-220, 302, 838, 303, 389, 408, 433, 452, 458.
GENERAL INDEX.
531
Refuse-piles, 200, 201.
Sinkers, perforated and grooved,
Religious ideas, 20-24, 410, 430.
337-340.
Reynolds, Sir Joshua, 127.
Skins, preparation of, 62.
Ribas, 362.
Skull of modern Indian, 160; of
Ribuult. Captain, 87, 229, 299, 330.
mound-builder, 160 161.
Rock-walls, 207-209.
Slave-posts in chunky-yards, 179.
Rook-writing, 62.
Sling-stones, 371.
Roe-deer," stalking of the, 323.
Slung-shots, 371, 372.
Roguet, 464.
Smith, Capt. John, 76, 91, 230, 260,
Romans, Captain Bernard, 6, 46, 83, Smoking, 393-399, 410.
92, 112, 256., 344.
Smoothing- stones, 292, 293.
Rosaries, 236.
Southern. Indians, physical charac
teristics of the, 8.
Spades, stone, 302.
Spanish invasions, effect of the, upon
Sacrifice of the firs-t-born male child,
the Indian population, 177.
13.
Spears used in capturing fish, 32S,
Salt, manufacture of, 45, 4G.
334.
Savannahs, the, 2.
Spear-heads, 240, 252,253; typical
Saws, 292.
forms of, 253,254; how made,
Scenauki, 69.
259-265.
>
Schoolcraft, H. R., 260, 439.
Spindle- whorls, 235.
^
Scrapers, 288-290, 496.
Spinning, 87.
Sculptured rocks, 377-380.
Spiral fire, 15.
\ Secondary interments, 103, 126, 131, Spoons, 500.
\
145, 160.
Sqnier, lion. E. George, 232.
Seminoles, the, 2, 4.
Sqnier and Davis, Messrs., 318, 354.
Sepulchral mounds, 183.
Stalling' s Island, shell-mound on,
Shawls, 61, 73.
197.
Shawnoes, the, 7.
Statues, wooden, at Talomeco, 25.
Shea, John Gilmary, 387.
Stones for rounding arrow-shafts,
Shell drinking-cups, 233.
366, 367.
Shell-heaps, 200, 201, 483.
Stones upon which nuts were crack
Shell-mounds, 195-200.
ed, 315-320.
Shell ornaments, 162.
Stone graves, 214-238; in Kacoo-
Shell pins, 233, 518.
chee Valley, 214; age of, 238; in
Shells, primitive uses of, 495-519 ;
the environs of Keowe, 216; in
as gouges, chisels, scrapers, and
Tennessee, 216-221; in Missouri,
knives, 496-498; as drinkitig-cups,
217; in Illinois, 218-220; in Eu
498-500 ; as spoons, 500 ; as ag
rope, 224.
ricultural implements, 500; as Stone heaps. 114, 202.
rattles, 500; as shrines for idols, Stone Mountain, 207, 380.
500; as an element of strength in Stone tumulus near Sparta, Georgia,
the manufacture of earthen-ware,
148.
501; as money, 501-511 ; trade Stonoes, the, 2.
in, 506; as ornaments, 511-519; Storehouses, 12, 308.
heads of, 511; typical forms of Stung-Serpcnt, funeral obsequies of
shell beads, 512 ; how perforated,
the, 105-107.
513, 514; ear-rings, 515 ; gorgets, Successive inhumations, 193.
515--517; armlets and anklets, Summer -houses, 35.
517; pins, 518; unfinished shell Sun, office of the, among the Natcb-
beads, as articles of cormnerce,
ez, 23, 24; worship of the, 20, 21,
519.
422, 427--429 ; truncated pyramids
Shoes, 77, 79, 81.
erected in honor of the, 22.
Sick, treatment of the, 28-33, 362- Supreme Berns?, conceptions of a,
364.
Silver Bluff, 123, 148.
Sword, stono, 268.
533
T.
Tacitus, 341. Talomeco, mausoleum of, 25, 230.
Tambours, 91.
Tattooing, 75, 80. Temple of the Natehez, 427-429. Temple-mounds, 138, 142, 158. Terra-cotta, vessels of, 454-466.
v Terron, Juan, 473. Timberlake, Lieutenant, 251, 285, 346, 419. Time, how reckoned, 100, 1C1. Toalli, dwellings of, 35.
Tobacco, 44, 393-399.
Tomahawks, 277. Tombs of the Virginia kings, 26, 27, Tomo-chi-chi, 69, 131, 185, 188, 189,
421. Tonti, the Chevalier, 24=3.
Town-plantation, 41. Towns of the Florida. Indians, 35,
37; of the Virginia Indians, 36; of the Carolina1 tribes, 37, Trade relations, 63, 64, 162, 238,
243, 506. Tribal or family mounds, 189. Tritui-ating-stones, 81*. Troost, Professor, 216, 438. Tubes, stone, 359-365. Tumlin, Colonel Lewis, 1&6. Tumuli, ancient, in Georgia, 121;
Bavtrarn's account of, 123-125 -, secondary uses of, 126, 160; gen eral distribution of, 127-129 ; as sociated in groups, 129 ; shapes and sizes of, 129, 130; few built, after the advent of the Europeans, 130-132; age of, 131-135; on the Etowah River, 136-143; in the valley of Little Shoulder-bone Creek, 143--147; on the Savannah River below Augusta, 153--157; on the Ocmulgee River, opposite the city of Macon, 158--161; on Lamar'a plantation, ,,162; on Brown's Mount, 165; onMessier's plantation,, 166-174; on Woolfolk's plantation, 182; sepulchral, 183; chieftain, 183; family or tribal, 189; on the low grounds of the Rivanna, 103; of shell, 195-200 ; of stone, 202 ; on Stai ling's Island, 197; in N~acoochee
Valley, 213; at the confluence
of the Etowah and Oostenaula Kivers, 253.. Tuscaroras, the, 7. Tydeus, 120. Tylor, Mr., 414.
TJchees, the, 2, 3. Undertakers, 112, 191, 223. Uppow6c, 396.
V.
Vaca, Cabeca de, 229, 245, 362. Venegas, Mignel, 363. Venetian beads, 235. Victory-s tones, 285. Virginia kings, how entombed, 108.
W.
Walled towns, 209-212. Walnut-oil, 45, 315, 316. Wampum, 501-511. War, how declared, 16, 18; conduct
of the Southern Indians in, 17, 18. War-chief, the great, dignity and
power of, 11, 16; represents the Hico in his absence, 16. Warriors, cabin of the, 16; charac teristics of the Southern, 19. Wears, 330-332. Weaving, 78, 87. Wedge-shaped axes, 278--281. Westoes, the, 2. Whetstones, 277, 367. Whittlesey, Colonel Charles, 319, 881. Widows, the care of, 13. Williams, Roger, 507. Wilson, Dr. Daniel, 384. Winter houses, 35. Wislizenus, Dr. A., 218. Woman-chief, among the N"atchez. 23. Woman, position of, 70. Wrightsboro, ancient monuments near, 123. Wyman, Prof. Jeffries, 200.
Y.
Yamasees, the, 2, 3. Year, divisions of the, 100.
B E*