'Jerboa' in the relation of the face to the cranium. The palate in the bent Newstead skull (Plate XCVI., Fig. 3)—instead of being in a line with the base of the cranium as in a typical 'forest' horse, or forming with the cranial base an angle of about 80 as in certain Arab and Thoroughbred strains—forms an angle of 18.60 with the cranium. To what extent the face was bent downwards in Equus sivalensis it is difficult to say, but, judging by the imperfect skulls hitherto discovered, the deflection was probably between 180 and 200; while in the relation of the face to the cranium the Newstead skull probably closely agrees with Equus sivalensis, it very decidedly differs in the size of the cheek teeth. On the other hand, the Newstead horse very closely agrees in its teeth with Equus stenonis.

As it happens, the bent Newstead skull is almost identical in its dimensions and its teeth with a skull in the St. Petersburg Museum of a horse from the Kirghiz steppe, in which the face is strongly deflected. It may hence be assumed that the bent Newstead skull belonged to an ancient race allied to Equus sivalensis of the Indian Pliocene, and to one or more of the varieties of Equus steflonis widely distributed over Europe at the beginning of the Pleistocene period.

The Newstead horse, with a decided fronto-nasal prominence, is especially interesting, because it forms a connecting link between Equus sivalensis and certain strains of the modern English racehorse. In all probability in Echidna, the 'fiddle-headed' grand-dam of 'Stockwell,' the face was strongly deflected on the cranium. In 'Stockwell' the face forms an angle of nearly 140 with the cranium; in his descendant, 'Persimmon, the angle is about 120. In some broad-browed Oriental horses with a long, tapering face the deflection is probably as pronounced as in 'Stockwell,' but in broad-browed Arabs with a short face the deflection may be only 40 or 50, and in narrow-browed Arabs with a nearly straight profile and a fine muzzle it probably rarely exceeds 100. As in 'Stockwell' the face was long and narrow as well as deflected, it may be assumed that this famous racehorse had in part sprung from Oriental ancestors of the 'Siwalik' type and in part from narrow-browed ancestors of the 'plateau' type evolved in North Africa, i.e. in part from Prof. Ridgeway's Equus caballus libycus.

In addition to almost perfect skulls belonging to horses which fairly accurately represented 'forest,' 'plateau,' and 'Siwalik' types, Newstead yielded a considerable number of skulls in a more or less perfect state of preservation, which evidently belonged to cross-bred animals. Two of these skulls belonged to broad-browed horses with a decidedly dished but somewhat long face, i.e. to horses which may have been a blend of the 'steppe' and 'forest' varieties. Similar skulls have been recorded from Swiss lake-dwellings, and there is still an ancient race of horses characterised by a broad forehead and a concave profile in the vicinity of Schlettstadt, in Upper Alsace. It is conceivable that members of the ancient race now represented by the Schlettstadt horses, recently described by Dr. Max Helzheimer, were brought to Newstead by German auxiliaries. The majority of the remaining skulls belonged to coarse-headed animals, which may very well have been a blend of the 'Siwalik' and 'steppe' types.