354 0. C. Marsh — Recent Polydactyle Horses. 



figures 18 and 19. The ulna is complete, but the fibula is not. 

 There is no diastema between the premolars, and the third, 

 fourth, and fifth are like the molars. 



In the Upper Miocene of Oregon, the series is continued in 

 31iohippus, a genus established by the writer in 1874 for 

 several species found in this formation (this Journal, vol. vii, 

 p. 249). All are much larger than Mesohippus, and more 

 specialized in the skull and feet. The latter are shown in 

 figures 20 and 21. The premolars form a continuous series, 

 and three of them are of the molar type. This genus appears 

 to be the American representative of Anchitherium, von 

 Meyer (1844), but is somewhat less specialized. 



In the early Pliocene of this country, the genus Protohippus 

 of Leidy (1858) is especially abundant, and several species 

 about as large as a donkey are known. All have three func- 

 tional toes on each foot, but these are more nearly of a size 

 than the digits of the European Hipparion of Christol (1832). 

 In the later deposits of this formation, three-toed horses are 

 replaced by Pliohippus, named by the writer in 1874 (this 

 Journal, vol. vii, p. 252). This genus has many species, all 

 smaller than the horse, but with similar feet. A true Equus is 

 entombed in a higher horizon, but apparently became extinct 

 before America was discovered. 



The type specimens of all the genera of fossil horses pro- 

 posed by the writer up to 1876 were carefully studied by Prof. 

 Huxley in that year, and made the basis of one of his New 

 York lectures.* The diagram shown on page 355 was prepared 

 for that lecture under his direction, from specimens in the 

 Yale Museum collected and described by the writer. It gives 

 the main characters of the genera, and represents clearly the 

 evolution of the horse as then determined from American 

 specimens. 



Since then, the writer has made large collections of remains 

 of these and other genera and species, and the whole forms 

 by far the largest series ever brought together. An examina- 

 tion of this collection makes still more evident the line of 

 equine descent on this continent, and the genera since named, 

 and described {Eohipptis, Helohippus, and Epihippus), help 

 to complete the series. Many details are yet wanting, but the 

 writer hopes at an early day to bring all the facts together, and 

 make out as far as possible the genealogy of the horse. 

 New Haven, Conn., March 24th, 1892. 



* American Addresses, p. 88, London, 1877. 



