894 SIR V. BROOKE ON THE [Nov. 1 9, 



In this Table, showing the geographical distribution of the Cer- 

 vidae exclusive of the Elk and Reindeer, which have a circumpolar 

 range, the following are the most important facts observable. Out 

 of a total of 61 species 39 are confined to the Old and 22 to the 

 New World. Of the 39 Old-world species 33 are found in the 

 Eastern Palsearctic and Indian regions, the latter containing 22 

 species, a much larger number than that contained in any one of the 

 remaining four regions. 



Of the 22 species confined to the New World, 8 are found to the 

 north and 14 to the south of the Isthmus of Tehuantepec, the 

 division adopted by Mr. Sclater between the Nearctic and Neotro- 

 pical regions. These figures cannot, however, be taken as afford- 

 ing a just estimate of the relative degree of specialization of the 

 Cervidae in the Nearctic and Neotropical regions respectively. On 

 the one hand, the three species of Cariacus in the latter are undoubt- 

 edly immigrants from the former region ; and on the other hand, the 

 species contained in the subgenera Coassus and Furcifer are, as 

 far as the antlers are concerned, amongst the least-specialized 

 of existing species. 



Out of the eleven genera into which the Old-world Deer are di- 

 visible, there is only one {Damn) which has not a representative in 

 either the Paleearctic or Indian region. 



Owing to the exceeding paucity of Palaeontological material, any 

 attempt to account for the present differentiation and distribution 

 of the CervidBe must be considered as purely tentative ; and I venture 

 to put forward the following hypothesis, more to provoke investiga- 

 tion and criticism, which may possibly bring new and more correct 

 ideas to light, than from any great confidence that I feel in its 

 approximation to completeness. 



In order to avoid unnecessary recapitulation, and to indicate as 

 clearly as possible the interdependence of the facts and hypothesis, 

 I shall place the facts and theory in two columns, numbering the 

 former, and merely introducing the number attached to each fact 

 before the deduction or deductions which are drawn from it. 



Facts. Hypothesis. 



(1) So many characters are (1) A very high degree of 

 common to the Cervidae and specialization had been attained 

 Bovidae that it is very difficult by the common ancestor of the 

 to lay down definitions distinc- Cervidae and Bovidae, especially 

 tive of the families. in the structure of the skull, 



dentition, and limbs ; hence the 

 possibility of divergence of cha- 

 racter in its descendants was 



(2) Dremotherium, the earliest much restricted. (2) For a long 

 known Cervine form, was with- period the early ancestors of the 

 out antlers and tetradactyle. Cervidae, which were tetradactyle, 



