NO. 2086. 



SOME MAMMALS OF THE PLEISTOCENE— HAY. 



561 



the eight specimens measured. The cranio-cephalic index has an 

 average of 33 and a fluctuation of 4.6, which is equal to 14 per 

 cent of the mean; while the facio-cephalic index has a variation of 

 only 6.2 per cent of the mean. It is somewhat remarkable that the 

 cranio-cephalic index varies so little in the broad-skulled group and 

 so much in the narrow-skulled horses. Now, each of these two 

 groups has all the characteristics of structure and proportion of a 

 well-defined species, such as the Przevalsky horse, Grevy's zebra, and 

 the chigetai. There appears to be no necessary relation between a 

 short skull and a wide forehead ; and therefore it can not be said that 

 the results are due to having selected the shorter skulled specimens. 

 The eight skulls of Grant's zebra have the average of the basilar 

 length exactly the same as that of the eight small domestic horses, 

 but the frontal width is considerably less and the cranio-cephalic 

 index, too, is less. We need not, therefore, believe that the pecul- 

 iarities of the two groups is due merely to size ; nor can we sup- 

 pose that the two groups are pure strains. We know that in each 

 there is some admixture of the one with the other ; and there may be 

 some mingling of possibly a third constituent. 



The writer believes, therefore, that it will be necessary to recog- 

 nize at least two distinct species among the progenitors of our 

 domestic horses. An objection to this view may be found in the fact 

 that these have bred and do breed freely together ; but this objection 

 will apply quite as well against our regarding Equus przevalskii as 

 a distinct species. One of the two supposed species is represented 

 in our time by the large narrow-faced horses, such as form the second 

 portion of the table on page 559 ; the other is represented by the pony- 

 like broad-faced horses, especially the Celtic pony and the fjord 

 horses of Norway. It has been already stated on page 552 that Ewart 

 in his work in the Transactions of the Edinburgh Royal Society 

 has presented certain measurements of a Celtic pony and that esti- 

 mates showed that its basilar length was close to 437 mm. Accept- 

 ing this and other measurements given by Ewart, likewise those of a 

 Celtic pony from the Loffoden Islands furnished by Stejneger, there 

 is obtained the following table : 



Measurements in millimeters and indices of two Celtic ponies. 



Specimen. 



Basilar 

 length. 



Vertex 

 length. 



Frontal 

 width. 



Cephalic 

 index. 



Cranial 

 length. 



Cranio- 

 cephalic 

 index. 



Facial 

 length. 



Facio- 

 cephalic 

 index. 



Ewart's 



437 

 456 



485 



192 

 210 



43.7 

 46.3 



170 

 155 



36.7 

 34.0 



333 

 336 



76.2 



Stejneger's 



73.7 









The specimen measured by Stejneger has an extraordinarily broad 

 face. It is broader, in fact, than either of the skulls of Ewart's two 

 59758°— Proc.N.M.vol.48— 14 36 



