364 Bulletin American Musevm of Natural History. [Vol. XXVII, 



(2) In both groups, according to Granger (quoted by Sinclair, 1908, 

 p. 77), the base of the coronoid process shows a pecuhar perforation which is 

 a separated portion of tlie dental canal. 



The idea that the Eocene Hyracoids may be related to the Artiodactyls, 

 which might be suggested by the general similarity of the upper and lower 

 molar teeth of Megalohijra.r to those of Ancodus gorringei (rf. Andrews, op. 

 cit., pi. xviii, figs. 1-3), seems to be negatived by the differences in the pre- 

 molars, which in MegaloJn/rax are much more molariform than in any 

 Artiodactyl. 



In conclusion, out of all the divergent resemblances cited above, the 

 writer is inclined to regard those indicating remote affinitv with the Probos- 

 cidea as having the most weight, and this too in spite of the great differences 

 in the dentition ((/. j). 107). This conclusion appears to be strengthened 

 by the evidence for common ancestry of the Hyracoidea and Embrithopoda, 

 the latter in turn showing many underlying features (p. 366) with the Pro- 

 boscidea and Amblypoda. Finally the existence of so many "cross- 

 resemblances" between the Proboscidea and the Perissodactyla by way of 

 Hi/ra.v seem more consistent with the hypothesis that all these now very 

 divergent orders have been derived from a common protungulate stock, 

 than with the hypothesis that all the resemblances are due to convergent 

 evolution. 



V. The Embrithopoda (Barypoda). 



In discussing the affinities of Arsinoltherium, the gigantic Eocene pachy- 

 derm of the Fayftm, Dr. Andrews (1906, ]>. 63) says: "Comparison with the 

 Hyracoidea is very difficult on account of the small size of the modern 

 representatives of the grouj) and our complete ignorance of the skeleton of 

 the large extinct forms. In the early Hyracoidea there is a tendency for the 

 anterior teeth (incisors, canine, and premolars) to form a continuous series, 

 the anterior pair of incisors being more or less enlarged and separated from 

 one another in the middle line by a short interval, as in Arsinoltherium. 

 The molars are brachyodont, but their pattern is such that if there were an 

 increase in height accompanied by an increasing infolding of the outer wall, 

 they might give rise to a type of tooth very similar to the molars of Arsino'i- 

 thcrium, as was explained above. Since, however, these early brachyodont 

 Hyraces are contemporaries of Arsinoltherium, it is not suggested that they 

 are closely related; but the circumstance that they both seem to have origi- 

 nated in the same region lends some probability to the belief that they may 

 have had a common ancestry in the late Secondary or earliest Tertiary 

 period." Dr. iVndrews also refers the Embrithopoda, along with the 



