covered with a thin solution of glue. It is figured below, both side and back view, also from above, Figs. 24, e and f, and 26, e. There was also found a perfect awl or piercing instrument (Fig. 15, b), made of the left half of the right metatarsal bone of the elk (Cervus Canadensis). A part of the larger end crumbled a little after bringing it to the air. It is 127 inches long, and would have been an effective weapon to use in close combat. From a burial mound in Tennessee Dr. Jones took a needle, as he styles it, or piercing implement measuring 14 inches in length.* "It had" he states "been fashioned with great care from the tibia of the American deer, and was probably used for piercing leather." The larger number of the fragments of bone found in this Naples mound were those of the deer.

Mound No. 7 is within 60 feet of No. 6, on the west. It is now of an oblong shape, being longest from north to south, but this is due to the fact that the western half of it has been removed by the washing of the river. This mound was opened in one place to the original surface but nothing was found at the time, except fragments of bones split in the same manner as those in mound No. 6. Among these were fragments of the humerus, tibia, radius, ribs, and vertebra of the deer (Cervus virginianus. The humerus and femur of the wild turkey were found intact, also two femora of the beaver, the ulna of a large bird not identified,

and the ulna of a small feline probably the skunk. All these fragments were found mixed with ashes and pieces of charcoal, indicating plainly that they are the remains of a feast. No fragmentary bones were met with, and nothing to indicate cannibalism. Of the other mounds of this vicinity only one was explored. This was No. 15, which is about 60 feet in diameter and 6 feet high. A shaft about 8 feet square was sunk

(* Explorations of the Aboriginal Remains of Tennessee. By Joseph Jones (Smithsonian Contributions to Knowledge, vol. XXII, p. 61.)

to the original level, developing the fact that it was a burial mound, The bones were so greatly decayed as to render it impossible to preserve even any fragments of the skulls. Large numbers of fragments of pottery were found, but no whole vessels. The material and many of the markings resemble those of specimens, already referred to, from the old cemetery on the margin of the river, distant about 300 yards. Judged from the fragments, the majority of the vessels would hold about 2 gallons each. One, however, was not much larger than the half of a cocoanut-shell and of about the same shape, Fig. 21.

FIG. 22. Pottery fragments from mound near Naples, Ill.

It is of a dark color, about one-quarter of an inch thick, symmetrically inade, and ornamented with lines about the sixteenth of an inch in depth,

arranged as to form small lozenges about half an inch in length. The spaces so ornamented are four in number, large at the top and tapering to a point at the bottom, ending near the pointed end of the cup. The smooth intervening spaces form a four-pointed star, folded over the bottom of the cup, as shown in the cut. The preceding figure represents fragments of pottery from mound No. 15. The pieces a and c are each ornamented with figures stamped into the soft clay. One fragment from mound No. 6 was ornamented with a similar figure; also several fragments from the ancient cemetery above alluded to.

It is difficult to determine what was used to make this impression, but it must have been something of vegetable growth, serving as a natural stamp. From the character of the imprints on the fragment from mound No. 6, it seems that the stamp was somewhat pliable, as some of the figures seem distorted. The pottery must have been designedly broken and deposited in mound No. 15, as no whole vessels were found and not enough fragments of any one to restore it. From a careful examination of fragments from the mounds, as well as those from the ancient burial-place, it has been possible to determine the size and form of many of them. The most usual size held from 2 to 3 gallons, and was shaped as in the accompanying cut, restored from fragments found in mound 15.

FIG. 23. Pottery, from mound near Naples, Ill.

Fragments from mound No. 6 measure at the rim, one 9 inches and another about 6, while those from mound 15 give the following: 103

inches, 10, 91, 83, and 61; and specimens from the ancient burial-place give diameters of 20, 141, 12, 91, 8, and 5 inches, respectively. The beadlike dots around the margin of Fig. 23 indicate holes punched from the inside of the vessel at intervals of about three-fourths of an inch, made with the end of a round stick about the size of a lead-pencil. The dots upon the outside indicate corresponding elevations made by the point of the stick being pushed nearly through. This method of ornamentation was found upon specimens in mounds Nos. 6 and 15, in specimens from the ancient cemetery, and in fragments found on an old village site, 3 miles west of Winchester.

The theory of a uniform typical skull-form for all the nations of the New World presented by Dr. Morton in his great work Crania Americana, so ably seconded by Dr. Nott in Types of Mankind indorsed by Humboldt, and for a time acquiesced in by American ethnologists, first challenged by Professor Retzius in 1859,* and again by Dr. Daniel Wilson in 1862,† may now be considered to have been completely overthrown.

It may be safely said that examples of all the various forms which the mania for skull classification has distinguished may be found among the various tribes of the New World yet living, as well as in crania exhumed from the ancient burial-places of extinct tribes.

Although the material for generalization is yet scanty, the same may be affirmed of the mound-builders and stone-grave race. The plan here. tofore followed in attempting to establish a typical skull-form both for the modern Indian and the mound-builders is wholly unsatisfactory and fallacious. Let us assume an experiment exceeding in magnitude any. thing yet attempted in that line, for upon the theory adopted, the greater the number of skulls examined the more the probability that the typical form ascertained is the correct one; and if we find this experiment open to great sources of error which we have no means of eliminating, we may safely conclude that the attempts made upon a much smaller scale have failed to furnish us any reliable information. Let us take one thousand skulls, ten skulls from each of one hundred tribes scattered from Hudson's Bay to Patagonia. We tabulate all the various measurements, longitudinal, parietal, frontal, vertical, &c., and by this means strive to obtain a typical skull. It may be that not five of the whole number conform to this type, and it may be that each of these belonged to one tribe. In attempting thus to establish a typical skull-form for a hundred tribes we assume a fact which does not exist, viz, that there is an average uniformity in the skull-forms of the various hundred tribes, or, in other words, that the average variation of skull-forms is the same in all these tribes. Secondly, we assume that the ten skulls taken rep resent fairly the variations of skull-forms in the particular tribe. This may be true or it may not. For example, every skull obtained may, by *Smithsonian Report, 1859, p. 264. t Ibid, 1862, p. 240 et seq.

accident, belong to the brachycephalic type, while the skulls of the tribe may be equally divided between that form and the dolichocephalic. With these facts before us examine any table of measurements yet made of American crania and calculate the chances of error.

There are not in all the public and private collections fifty undoubted mound-builder skulls, assuming that the builders of the great mounds and earth-works were a different race from the modern Indian. With a view of ascertaining the probable number of such skulls in the country, in January, 1880, the author mailed a circular letter to a large number of private individuals and public institutions, about seventy in all, requesting answers to two questions:

1st. How many genuine mound-builder skulls are in your collection 2d. How many that are supposed to be mound-builder skulls ?

In a very large majority of the cases the answer to even the second question was "not one!" while the exception was rare, indeed, where persons claimed to be in possession of skulls of the first character. The four greatest public collections are the Academy of Science, Philadelphia; the Smithsonian Institution, at Washington; the Peabody Institute, Cambridge; and the Davenport Academy of Science, in Iowa. In these public collections are many labeled mound-skulls, simply from the fact that they came from the base of a mound, without any reference to its size or other articles taken from it. In estimating the number of undoubted mound-skulls, those of this character are excluded. Assuming a difference in the race who built the great mounds and earth-works of the Mississippi Valley and the modern Indian, a skull taken from a mound of this character may belong to the one or the other. Applying the test suggested in the circular above alluded to, which is the best the nature of the subject will admit of, that is, "to only class as genuine those that are found in connection with other objects that unquestionably belonged to the mound-builders," the number of genuine mound-builder skulls is reduced by two-thirds.*

Dr. Foster fell into a grave error when he classed as "authentic skulls of the mound-builders" those obtained from low mounds on the banks of the Des Plaines River, of Illinois. These mounds were elevated only about "2 feet above the surrounding plain," and no objects were found in the mounds which are looked upon as peculiar to the mound-builders. Yet he classes these skulls as genuine mound-builder remains, figures some of them in his work; and in the winter of 1869-1870 presented to

* Dr. J. F. Snyder, a gentleman who has devoted a great deal of thought to this subject, suggests that "this test appears very unsatisfactory, for the difficulty of distinguishing 'objects that unquestionably belonged to the so-called mound-builders' is necessarily as great as to distinguish their crania." The force of this criticism is admitted; yet if we find a skull at the base of a great mound, and with it copper axes, pipes with the peculiar curved base, and such articles as were found by Squier and Davis in the mounds of Ohio, we may safely assume that the mound was built by the same race that erected those of similar character in which similar articles were found.