Their arms were extended and their hands slightly elevated, as if together holding up a sandstone mortar which was between their faces.

This stone is somewhat hemispherical, about 2 feet in diameter across the top, which is hollowed in the shape of a shallow basin or mortar. It had been subjected to the action of fire until burned to a bright red.

The cavity was filled with white ashes, containing small fragments of bones burned to cinders. Immediately over this, and of sufficient size to cover it, was a slab of bluish-gray limestone about 3 inches thick, which had small cup-shaped excavations on the under side. This bore no marks of fire. Near the hands of the eastern skeleton were a small hemat.i.te celt and a lance-head, and upon the left wrist of the other two copper bracelets. At the depth of 25 feet, and on the natural surface, was found what in an Ohio mound would have been designated an "altar."

This was not thoroughly traced throughout, but was about 12 feet long and over 8 feet wide, of the form shown in Fig. 24.

It consisted of a layer of well-prepared mortar, apparently clay, slightly mixed with ashes. This was not more than 6 or 8 inches thick in the center of the basin-shaped depression, where it was about 1 foot lower than at the other margin. It was burned to a brick-red and covered with a compact layer of very fine white ashes, scattered thickly, through which were small water-worn bowlders, bearing evidences of having undergone an intense heat. Mingled with this ma.s.s were a few thoroughly charred human bones. The material of the shaft, after the first 3 feet at the top, consisted almost wholly of finely packed ashes, which appeared to have been deposited at intervals of considerable length and not at one time.

It is evident from this description, which is abridged from the report of the a.s.sistant, that we have here a true representation of the so-called "altars" of the Ohio mounds. But, contrary to the usual custom, as shown by an examination of the Ohio works, this mound appears to have been used by the people who erected it as a burial place, for the mode of construction and the material used for the body of it forbid the supposition that the lower burial was by a different people from those who formed the clay structure at the base.

It is proper to state that around and near the inclosure (No. 7 of Plate V) were a number of stone graves of the ordinary box shape, constructed in the usual way, of stone slabs.

At this place was also discovered a pit or cache resembling those found at Madisonville, Ohio. A more thorough examination will probably bring to light others.

The descriptions of other burial mounds of this region, differing slightly in minor details from those mentioned, might be presented, but the foregoing will suffice to give the types and show the character of the structures of this kind in this section. The details given will, I think, satisfy any one that the authors of these structures were also the authors of the Ohio works, or that they belonged to tribes so closely related that we may justly consider them as one people.

I have been and am still disposed to connect the mound-builders of the Kanawha valley with those of western North Carolina, but our explorations in the two sections have convinced me of their close relation to the people whose mysterious monuments dot the hills and valleys of Ohio. That they were related in some way to the mound-builders of North Carolina and East Tennessee is more than probable, but the key to unlock this mystery, if it exists anywhere, is most likely to be found in the history, traditions, and works of the Cherokees, and the traditions relating to the Tallegwi.

As a result of my examination and discussion of the burial mounds of Wisconsin, I reached the conclusion that they were built by the Indian tribes found inhabiting that section at the advent of the whites, or by their ancestors. The data, of which but a comparatively small portion is given, seem to justify this conclusion. But the case is somewhat different in reference to the works of the Ohio district. Although the data obtained here point with satisfactory certainty to the conclusion that Indians were the authors of these works, it cannot be claimed that all or even the larger portion of them were built by Indians inhabiting the district when first visited by the whites, or by their ancestors.

[Ill.u.s.tration: PL. VI. ENLARGED PLAN OF PART OF THE WORKS SHOWN IN PL.

V.]

Hence the mystery which enshrouds them is deeper and much more difficult to penetrate than that which hangs about the antiquities of some of the other districts; in fact, they present probably the most difficult problem for solution in this respect of any ancient works of our country. That some of the burial mounds, graves, and other works are to be attributed to Indians who entered this district after the Europeans had planted colonies in Canada and along the Atlantic coast is probably true, but that much the greater portion of the typical works belong to a more distant period must be conceded. It is a singular fact that in the latter half of the seventeenth century, when European explorers began to penetrate into this region, what is now the State of Ohio was uninhabited.

The Miami confederacy, inhabiting the southern sh.o.r.e of Lake Michigan, extended southeasterly to the Wabash. The Illinois confederacy extended down the eastern sh.o.r.e of the Mississippi to about where Memphis now stands. The Cherokees occupied the slopes and valleys of the mountains about the borders of what is now East Tennessee, North Carolina, and Georgia. The great basin bounded north by Lake Erie, the Miamis, and the Illinois, west by the Mississippi, east by the Alleghanies, and south by the headwaters of the streams that flow into the Gulf of Mexico, seems to have been uninhabited except by bands of Shawnees, and scarcely visited except by war parties of the Five Nations.[35]

With the exception of some slight notices of the Erie or Cat Nation dwelling south of Lake Erie, the mere mention of the Tongarias (possibly but another name for the Eries, with whom Colden identifies them), located somewhere on the Ohio, and the tradition regarding the Tallegwi, the only history which remains to us regarding this region previous to the close of the seventeenth century, is to be gathered from the ancient monuments which dot its surface. Even conjecture can find but few pointers on this desert field to give direction to its flight. But it does not necessarily follow, because we are unable to determine the direction in which the goal we are seeking lies, that we cannot tell some of the directions in which it does not lie, and thus narrow the field of our investigation. I will therefore venture to offer the following suggestions:

As the evidence in regard to the antiquities of the northwestern, the southern, and the Appalachian districts points so decidedly to the Indians as the authors, I think we may a.s.sume that the works of Ohio are attributable to the same race. As they bear a strong resemblance in several respects to the West Virginia and North Carolina works, and as the geographical positions of the defensive works indicate pressure from the north and north-west, we are perhaps justified in excluding from consideration all tribes known to have had their princ.i.p.al seats north of the Ohio in historic times, except the Eries, which form an uncertain and so far indeterminable factor in the problem.

The data so far obtained seem to me to indicate the following as the most promising lines of research: The possible ident.i.ty or relation of the Tallegwi and the Cherokees; the possibility of this region having been the ancient home of the Shawnees or their ancestors (though I believe the testimony of the mounds is most decidedly against this and the following supposition); and the theory that the builders of these works were driven southward and were merged into the Chahta-Muscogee family.

Be our conclusion on this question what it may, one important result of the explorations in this northern section of the United States is the conviction that there was during the mound-building age a powerful tribe or a.s.sociation of closely allied tribes occupying the valley of the Ohio, whose chief seats were in the Kanawha, Scioto, and Little Miami Valleys. We might suppose that one strong tribe had occupied successively these various points, yet the slight though persistent differences in methods and customs indicated by the works seem to favor the other view. Moreover, the data furnished by the burial mounds lead to the conclusion that all the works of these localities are relatively contemporaneous. Not that those of either section are all of the same age, perhaps by some two or three or possibly more centuries, but that those of one section, as a whole, are relatively of the same age as those of the other sections. Nevertheless a somewhat careful study of all the data bearing on this subject leads me to the conclusion that the Cherokees are the modern representatives of the Tallegwi, and that most of the typical works of Ohio and West Virginia owe their origin to this people.

In each section there are some indications that the authors of these works followed the custom of erecting burial mounds down to the time the Europeans appeared on the continent. These evidences have not been given here, as it is not my intention to discuss them in this paper.

In Ohio there are undoubted evidences of one, if not two, waves of population subsequent to the occupancy of that region by the builders of the chief works. But these were of comparatively short duration, and were evidently Indian hordes pressed westward and southward by the Iroquois tribes and the advance of the whites.

THE APPALACHIAN DISTRICT.

This district, as already defined, includes East Tennessee, western North Carolina, southwestern Virginia, and the southeastern part of Kentucky. It is probable that northeastern Georgia and the northwestern part of South Carolina should be included, but the investigations in most of the sections named have not been sufficiently thorough to enable us to fix with any degree of certainty the boundaries of the district.

Although there is uncertainty in reference to the area occupied by the people who left behind them the antiquities found in this region, there can be no doubt that here we find a cla.s.s of burial mounds differing in several important respects from any we have so far noticed.

Some of the most important mounds of this cla.s.s found in this district were discovered in Caldwell County, North Carolina, and opened in 1882 by Mr. J. P. Rogan, one of the Bureau a.s.sistants, aided by Dr. J. M.

Spainhour, a resident of the county.

As Mr. Rogan's descriptions are somewhat full, I give them substantially as found in his report:

_The T. F. Nelson mound._--This mound, so insignificant in appearance as scarcely to attract any notice, was located on the farm of Rev. T. F.

Nelson, in Caldwell County, North Carolina, on the bottom land of the Yadkin, about 100 yards from the river-bank. It was almost a true circle in outline, 38 feet in diameter, but not exceeding at any point 18 inches in height. The thorough excavation made revealed the fact that the builders of the mound had first dug a circular pit, with perpendicular margin, to the depth of 3 feet, and 38 feet in diameter, then deposited their dead in the manner hereafter shown, and afterwards covered them over, raising a slight mound above the pit.

A plan of the pit, drawn at the time (after the removal of the dirt), showing the stone graves and skeletons, is given in Fig. 25.

The walled graves or vaults and altar-shaped ma.s.s were built of water worn bowlders and clay or earth merely sufficient to hold them in place.

No. 1, a stone grave or vault standing exactly in the center of the pit.

In this case a small circular hole, a little over 3 feet in diameter and extending down 3 feet below the bottom of the large pit, had been dug, the body or skeleton placed perpendicularly upon its feet, and the wall built up around it from the bottom of the hole, converging, after a height of 4 feet was reached, so as to be covered at the top by a single soapstone rock of moderate size. On the top of the head of the skeleton and immediately under the capstone of the vault were found several plates of silver mica, which had evidently been cut with some rude implement. Although the bones were much decayed, yet they were retained in position by the dirt which filled the vault, an indication that the flesh had been removed before burial and the vault filled with dirt as it was built up.

[Ill.u.s.tration: FIG. 25.--Appearance of T. F. Nelson mound after excavation.]

Nos. 2, 3, 4, 5, 6, 7, 8, 9, and 10, although walled around in a similar manner, were in a sitting posture on the bottom of the pit. In the grave of No. 2 was found a polished celt, in that of No. 3 a single discoidal stone, in that of No. 6 two polished celts, and immediately over No. 9 a pitted stone.

Nos. 11, 12, and 13 are three skeletons in a squatting posture, with no wall around them and unaccompanied by relics of any kind.

Nos. 14 and 15 are two uninclosed skeletons, lying horizontally at full length. With the former some pieces of broken soapstone pipes were found, and with the latter one polished celt.

No. 16, an uninclosed "squatter," of unusually large size, not less than 7 feet high when living. Near the mouth was an uninjured soapstone pipe.

The legs were extended in a southwest direction, upon a bed of burnt earth.

The faces of all the squatting skeletons were turned away from the standing central one.

At A was found a considerable quant.i.ty of black paint in little lumps, which appear to have been molded in the hull of some nut. At B was a cubical ma.s.s of water-worn bowlders, built up solidly and symmetrically, 24 inches long, 18 inches wide, and 18 inches high, but with no bones, specimens of art, coal, ashes, or indications of fire on or around it.

Many of the stones of the vaults and the earth immediately around them, on the contrary, bore unmistakable evidences of fire; in fact, the heat in some cases left its mark on the bones of the inclosed skeletons, another indication that the flesh had been removed before burial here, either by previous burial or otherwise.

Scattered through the dirt which filled the pit were small pieces of pottery and charcoal. The bottom, and sides of the pit were so distinctly marked that they could be traced without difficulty.

This mound stood about 75 yards south of the triangular burial pit described below.

[Ill.u.s.tration: FIG. 26.--Burials in the T. F. Nelson triangle, Caldwell County, North Carolina.]

_The T. F. Nelson triangle._--This is the name applied by Mr. Rogan to an ancient triangular burying ground found on the same farm as the mound just described and about 75 yards north of it.

It is not a mound, but simply a burial pit in the form of a triangle, the two longest sides each 48 feet and the (southern) base 32 feet, in which the bodies and accompanying articles were deposited and then covered over, but not heaped up into a mound; or, if so, it had subsequently settled until on a level with the natural surface of the ground. The apex, which points directly north, was found to extend within 3 feet of the break of the bank of the Yadkin River, the height above the usual water-level being about 12 feet. The depth of the original excavation, the lines of which could be distinctly traced, varied from 2-1/2 to 3 feet. A rude sketch of this triangle, showing the relative positions of the skeletons, is given in Fig. 26.

Nos. 1, 2, 3, 4, 5, 6, 7, 8, and 9 indicate the positions of single skeletons found lying horizontally, on their backs, heads east and northeast. With No. 2 was found a broken soapstone pipe, and with Nos. 5 and 9 one small polished celt each.

Nos. 10, 11, 12, 13, 14, and 15 indicate the positions of skeletons inclosed in rude stone vaults built of cobblestones and similar to those in the preceding mound. (See Fig. 25.) Nos. 10, 12, 13, and 15 were in a sitting posture, without any accompanying articles.

Graves 11 and 14 contained each two bodies, extended horizontally, the lower ones, which were of smaller stature than the upper ones, face up and with heavy flat stones on the extended arms and legs. The upper ones, with face down, were resting on those below. No implements or ornaments were found with them.

Near No. 12 about a peck of singular, pinkish-colored earth was found.

[Ill.u.s.tration: FIG. 27.--Engraved sh.e.l.l gorget from mound, Caldwell County, North Carolina.]

In the northwest part of the triangle (at A in Fig. 26) ten or more skeletons were found in one grave or group, which from the arrangement the explorers concluded must have been buried at one time; the "old chief" (?), or princ.i.p.al personage of the group, resting horizontally on his face, with his head northeast and feet southwest. Under his head was a large engraved sh.e.l.l gorget (Fig. 27); around his neck were a number of large-sized sh.e.l.l beads, evidently the remains of a necklace; at the sides of the head, near the ears, were five elongate copper beads, or rather small cylinders, varying in length from one and a quarter to four and a half inches, part of the leather thong on which the smaller were strung yet remaining in them. These are made of thin pieces of copper cut into strips and then rolled together so that the edges meet in a straight joint on one side. (See Fig. 28.) The plate out of which they were made was as smooth and even in thickness as though it had been rolled.