In preparation the skull and jaws have not been separated from their position in the matrix, so that the triturating surfaces of the teeth have not been exposed. The specimen shows interesting pathological structures in the shape of battle scars which will be discussed in a subsequent paragraph on the habits of entelodonts. Archæotherium mortoni Leidy. Five skulls of small entelodonts secured by the Princeton 1920 Expedition from the "turtle-oreodon layer" of the lower Oreodon beds in the drainage basins of Indian and Bear Creeks in Pennington County, South Dakota, agree so closely with Mr. Troxell's A. clavus clavus that, with his description before me, I can not separate our material from his. Another specimen, No. 11009, from the lower Oreodon beds in Corral Draw, collected by Mr. H. F. Wells in 1894, shows remarkably close agreement with A. clavus darbyi Troxell, the few differences between them not being of specific rank in my opinion. None of this material, however, have I been able to separate from A. mortoni. Although the type of this species is a fragment of the left side of the face, with p3 and p4 in place, more complete skulls, originally in the collection of Dr. Owen, were later referred by Dr. Leidy himself to A. mortoni and beautifully figured,' constituting "heautotypes "s of the first order of importance. These specimens lack both dependent and posterior jugal processes, but otherwise agree closely with the Princeton material, allowing a certain amount of variation in size for difference in age, sex and individuals. Whatever be the ultimate status of Mr. Troxell's species, I am disposed to refer all of our small entelodonts from the Big Badlands to Leidy's A. mortoni. Mr. Troxell's figured specimen of A. clavus clavus lacks the tip of the cheek flange, and it is also indicated as missing in the specimen figured by Peterson and referred to A. mortoni. As this is complete in two of our Princeton skulls, I have illustrated it in the accompanying figures (Figs. 7, 8). Figured by Dr. Leidy in "Ancient Fauna of Nebraska," Pl. IX., Fig. 3. 7" Ancient Fauna of Nebraska," Pls. VIII., IX. | ་་ 8 Schuchert and Buckman, Science, n. s., Vol. 21, No. 545, p. 900, 1905. 9"A Revision of the Entelodontidae," Memoirs Carnegie Museum, Vol. IV., No. 3, Figs. 4-6, May, 1909. In No. 12529, found last summer by Mr. Wanless in the clays of the "turtle-oreodon layer" of the lower Oreodon beds in the valley of Indian Creek, Pennington County, South Dakota, a young skull FIG. 7. Archæotherium mortoni Leidy. No. 12529. Portion of left jugal seen from the outer side showing both the dependent and the posterior processes, one half the natural size. FIG. 8. Archæotherium mortoni Leidy. No. 11440. End of the dependent process of the right jugal showing concavity of anterior margin and thickening of distal end, one half the natural size. with both milk and permanent premolars in place and the third molar just appearing, the process is slender and thin, 8 mm. in maximum thickness in the middle and about two millimeters thicker at the distal end. Transversely, it is 41 mm. at the widest part. In the second specimen, No. 11440, from the lower Oreodon beds at the head of Sand Creek, Nebraska, collected by Mr. J. W. Gidley in 1896, the tip of the process on the right side seems to have been injured during life, for the anterior margin is concave distally and the process ends in an expanded area with a rugose flattened face. measuring 15 mm. vertically by 27 transversely. No comparison can be made with the corresponding structure on the opposite side, as this has been broken off. A. mortoni is represented by a larger number of specimens than any other entelodont in the Oreodon beds, especially in the zone of rusty caliche nodules in the upper part of the "red layer," from which came all the specimens of this species collected by the Princeton 1920 Expedition. Probably it will also be found to range down into the Titanotherium beds, but has not been identified with certainty in the Princeton collections from that horizon. Apparently it does not occur in the Protoceras beds. Lower Jaws. That there were at least two entelodonts of large size in the present area of the Big Badlands during the deposition of the turtle-oreodon layer" of the lower Oreodon beds is shown by two lower jaws secured by the Princeton 1920 Expedition in the valley of Indian Creek in Pennington County. The larger of these, No. 12550, retains the mental processes and agrees so closely in size and configuration of the jaw with the type of A. wanlessi that I do not hesitate to refer it to that species. The incisors, tips of the canines and crowns of p3 have been lost, but otherwise the dental series is complete (Fig. 9). The first premolar, measuring 21 mm. antero ※ FIG. 9. Archæotherium wanlessi sp. nov. No. 12550. Crown view of the lower teeth of the left side, one third the natural size. posteriorly, is double-rooted, but the roots are so close together that the empty alveolus would probably be indistinguishable from that of a single-rooted tooth. The anterior premolars are not as widely spaced as in the type (c-p1, 6.75 mm.; p1-p2, 6.75 mm.; p2-p3, 5 mm.). Pź is a stout tooth with slight anterior and posterior cingula and a blunt cutting edge front and rear. P4 has a broad mammillated heel narrowing posteriorly and a strong anterior cingulum. The most peculiar feature of all is the presence of a single cusp, the hypoconid, on the talonid of m3, internal to which there is a mammillated cingulum-like edge with hypoconulid and entoconid undifferentiated from the rest of this structure, except for a small swelling which perhaps corresponds to the entoconid. Unfortunately, these observations have to be based on a single tooth, for the corresponding molar on the right side has the crown badly shattered. There is less marked contrast in length and width between mī and m2 than in the second lower jaw to be described, No. 12546, which, unfortunately, has lost the mental processes, but agrees closely with the dimensions given by Mr. Troxell for the paratype of Marsh's Archæotherium crassum, as may be gathered from the table of measurements given below. In this specimen (Fig. 10) pī is FIG. 10. Arthæotherium crassum? (Marsh). No. 12546. Crown view of the lower teeth of the left side, one third the natural size. definitely double-rooted, 10 mm. back of the canine and the same distance from p2, which is 10-12 mm. from p3, and the latter 4 mm. from p4, which carries a broad mammillated heel, tapering in width posteriorly. The anterior molar seems proportionately smaller in comparison with the tooth back of it than in the other specimen, there is an external cingulum about the hypoconid in m2, and m3 has the normal heel development with two major cusps and a mammillated, cingulum-like hypoconulid. There is also close agreement in size with Leidy's fragmentary type of A. robustum,10 where the hypoconulid in m3 is a little stronger than in the Princeton specimen. The table of measurements below shows a close approximation to crassum and a rather wide departure from mortoni, and, in view of the extremely fragmentary character of the type of robustum, later referred by its author to mortoni, I am inclined to identify our specimen, provisionally, as A. crassum. Still another form, perhaps to be referred to Leidy's A. ingens, is represented by a fragment of mandible in the Princeton collection, No. 10875, from the lower Oreodon beds in Corral Draw, South Dakota, where the single anterior mental process preserved is of the ingens type. 10" Ancient Fauna of Nebraska," Pl. X., Figs. 12, 13. III. FROM THE PROTOCERAS BEDS. Scaptohyus altidens gen. et sp. nov. Type No. 11161, Princeton University Geological Museum, front This remarkable entelodont has figured several times in the litera- |