112 THE MAMMALIA OF THE DEEP RIVER BEDS. 



The facts which exploration is continually bringing to light tend more and more 

 strongly to the confirmation of Schlosser's view. " Derselbe [d. h. der Pferdestamm] 

 hat schon friihzeitig Eepresentanten in Europa sowohl als auch in Nordamerika, doch 

 sind nnr die neuweltlichen Glieder dieses Stammes von wesentlicher Bedeutung, 

 indem die altweltlichen sammtlich friiher oder spater ohne Hinterlassnng von ISTach- 

 kommen wieder ausgestorben sind. Die europaische Reihe erganzte sich immer 

 wieder durch Einwanderung amerikanischer Typen. Erst vom Pliocan an scheint 

 der Pferdestamm in der alten Welt weiter entwicklungsfahig geworden zn sein " 

 (No. 30, p. 486).* That the genus AncTiiilierium itself, even in the restricted sense 

 in which I have used that term, is of American rather than European origin is ren- 

 dered probable by the following considerations: (1) In the Old World the equine 

 series is very fragmentary and incomplete ; between JPachynolophus of the upper 

 Eocene and Ancliitlierium of the upper Miocene there is a great gap, which no known 

 European genus tends to bridge over, for assuredly Ancliilophus cannot be considered 

 in this connection. The three descending stages in the phylum, which we call JSpi- 

 hijjpus, Mesolup)pus and Mioliippus, have as yet yielded no representatives at all in 

 Europe, and even should one or other of them be found there hereafter, it is not in 

 the least likely that such a wealth of individuals and species as characterizes the 

 various horizons of the upper Eocene and the Miocene in this country will be discov- 

 ered in the Old World. That the line should be thus broken in the Eastern and 

 uninterrupted in the Western Hemisphere is surely a strong indication that the latter 

 was the theatre of its development, especially in view of the abundance of both indi- 

 viduals and species. (2) There is little difficulty in deriving Ancliitlierium from some 

 of the species of Miohippus ; the changes involved are slight, though some of them 

 are of much morphological significance, (a) In the first place, there is great increase 

 in size, both of the known species of this genus much exceeding any known form of 

 Miohippus. (jb) In the upper molars and premolars the conules are reduced in rela- 

 tive importance and the posterior transverse crest has become connected with the 

 outer wall of the crown, (c) In the lower premolars the internal cusps (a, a\ b, i\ 

 of Piitimeyer) are likewise reduced, and in more or fewer of these teeth the pillars, 

 anterior and posterior, are obsolete. A similar tendency may be observed in some 

 forms of MioMppus. (d) The odontoid process, which in the John Day genus is just 

 beginning to assume the spout-like shape, has in Ancliitlierium become as completely 

 so as in the horse, (e) The median digit of both mantis and pes has become greatly 

 enlarged and thickened, though there is no great reduction of the lateral digits, and 



*I believe Mine. Pavlow Las expressed a similar opinion as to the position of Ancliitlierium, though I cannot 

 lay my hand upon the reference. 



