378 E. L. Troxell — Entelodonts in Marsh Collection. 



or dentine sheath around the roots of the anterior teeth 

 back to P 4 . It greatly enlarges the size of the root up to 

 the enamel border. In one place on the left P 2 posterior, 

 this sheath ended very bluntly or else actually covered a 

 part of the enamel. 



The two sets of tubercles are subequal in length ; those 

 anterior are heavier, while those posterior are compressed 

 dorso-ventrally. It seems probable that their great lat- 

 eral extent may be correlated with the wide spread of 

 the lower teeth. Each of the tubercles is twisted upward 

 and forward, and in each case the dorso-posterior corner, 

 bent upward, is the longer. The chin is flat or slightly 

 concave transversely. 



The anterior mental foramen is not large; it lies 

 directly over the anterior tubercle and under the front 

 part of Pj. Another small foramen may be seen under 



ArchcBotherium crassum, specimen No. 10037. — This 

 third specimen has given rise to much speculation as to 

 whether it is or is not a part of the holotype of A. crassum. 

 The larger part of ninety or more fragments have been 

 matched together to form a portion of the lower jaw, a 

 maxillary and the dependent process from the jugal pre- 

 senting a most intricate problem in their preparation. 



The specimen is somewhat larger than A. marshi (fig. 

 10), but has a double-rooted first lower premolar. It 

 came from Gerry's Ranch, northeastern Colorado. 



Summary of Archceotfoerium crassum Group. 



A. crassum (Marsh): holotype, Cat. No. 12020; asso- 

 ciated specimens, Cat. Nos. 10036, 10037, 10288. 



Specimen No. 10037, alliance based on: (1) great re- 

 semblance of form of dependent process ; (2) similarity of 

 size; (3) found at the same time, 1870, and place, north- 

 eastern Colorado; (4) possibility of its being the same 

 individual because of no duplication of parts, similarity 

 of material itself, as well as form of dependent process, 

 and because Marsh borrowed the measurement from this 

 symphysis for his description of the holotype. 



Specimen No. 10036 is associated with A. crassum be- 

 cause of : (1) same accession number as the holotype, and 

 probably same locality, (2) similarity of form to rami of 

 No. 10036, (3) Marsh's description of it as the paratype. 



