888 SIR V. BROOKE ON THE [Nov. 19, 



racter lately discovered by Professor Garrod (see P. Z. S. 1877, p. 13, 

 fig. 24) has confirmed me in this opinion. 



Professor Garrod has observed that in all the American Deer, with 

 the exception of Cervus canadensis, the vertical plate developed from 

 the lower surface of the vomer is prolonged sufficiently downwards 

 and backwards to become ankylosed to the horizontal plate of the 

 palatals, and to form a complete septum dividing the nasal cavity 

 into two distinct chambers. The posterior edge of the vertical plate 

 of the vomer is plainly visible upon looking into the posterior nares 

 in the dry skull. In none of the Deer confined to the Old World is 

 tbe vertical plate of the vomer sufficiently developed to reach the 

 horizontal plate of the palatals ; and therefore in none is there a sep- 

 tum dividing the nasal cavity into two chambers separated from the 

 anterior to the posterior nares. 



This cranial character agrees with the pedal characters in placing 

 the Reindeer, one of the species of circumpolar distribution, with the 

 Deer of the New World. Alces, along with Hydropotes and Ca- 

 preolus, all of which are Telemetacarpi, Prof. Garrod's cranial 

 character places with the Plesiometacarpi of the Old World. 



So long ago as 1836, Dr. Gray 1 pointed out characters afforded by 

 the tufts of hair observable on the tarsus and metatarsus of most 

 Deer, and expressed his opinion that they constitute "a means of sub- 

 dividing them in natural sections." In his Catalogue of the Ruminant 

 Mammalia in the British Museum (1872), Dr. Gray divides the 

 Cervidae into two primary groups : — ( 1) that in which tuft of hair on 

 the outside of the metatarsus is above the middle of that bone, and 

 (2) that in which this tuft is below the middle of the metatarsus. 



Now I cannot but think it is a highly significant fact, and one 

 that should make us very cautious in condemning any character as 

 trivial, to find that a classification based on these apparently super- 

 ficial characters coincides exactly, as far as it goes, with that founded 

 on Prof. Garrod's cranial characters, and, with the exceptions referred 

 to, with that based on the osseous structure of the limb. 



There are some species both in the Old and the New World which 

 are devoid of either metatarsal or tarsal tufts. When present, the 

 metatarsal tuft is always above the middle of the metatarsus ; and 

 there is never any tuft on the inner surface of the tarsal joint in the 

 Deer confined to the Old World. In the Deer confined to the New 

 World, the metatarsal tuft when present is always below the middle 

 of the metatarsus, and there is frequently a distinct tarsal tuft. Cer- 

 vus canadensis constitutes the solitary exception. 



There is yet another character which lends some weight to the 

 teaching of those already mentioned ; and for the observation of it we 

 are again indebted to Prof. Garrod 2 . Prof. Garrod has observed that 

 the ascending rami of the prsemaxillse articulate with the nasals in 

 all the Old-World Deer, with one or two exceptions, whilst in the 

 Deer of the New World, again with one or two exceptions, the rami 

 of the praemaxillae do not reach the nasals. 



I shall not now attempt to decide the relative value of these charac- 

 » Vidfl P. Z. S. 1836, p. 66. - P. Z. 8. 1877, p. 13. 



