GEOGRAPHICAL DISTRIBUTION AxND DESCEiNT. 



^n 



brido-es for them, could the Polar bears be- 

 come simultaneously distributed over both 

 hemispheres. The similarity, and, in fact, 

 almost complete identity, of so many species 

 which are found, on the one hand, in the 

 Hudson's Bay Territories and in Canada, on 

 the other hand, in Siberia and Lapland, may 

 be appealed to as a proof of this connection 

 between the two hemispheres in the far 

 north. 



If these last lacts stand firm and irrefutable, 

 we possess, on the other hand, well-grounded 

 indications of less distant relationships. Dur- 

 ing the Miocene period the hyaenas were 

 distributed over a wide area in Southern 

 Europe; Sansans, Montpellier, Pikermi are 

 sign-posts pointing- to Asia, where the Sewalik 

 Hills at the foot of the Himalayas form the 

 boundary-mark on the east. The cave- 

 hysena {^Hycvna spclcra) inhabited in Quater- 

 nary times retreats in France, Germany, 

 Belgium, and even in England, which was 

 then connected with the Continent. But this 

 cave-hyaena presents so much resemblance to 

 the still living spotted hyaena (//. crociita) 

 that the latter might well be regarded as its 

 direct descendant, and, on the other hand, as 

 Gaudry has shown, possesses such close affini- 

 ties to a species found in the Pliocene deposits 

 ol Perrier in Auvergne (//. pci-ficri) that 

 this may be looked upon as its immediate 

 ancestor. The striped hyaena of the present 

 day is distinguished only by insignificant 

 details from another species likewise found 

 at Perrier (//. arvcrncnsis). The brown 

 hyaena may be regarded as the direct descen- 

 dant of a species found at Pikermi (//. cxiiuia). 

 The origin of the three species of hyaena now 

 living is accordingly pretty plainly demon- 

 strated. 



The study of the milk dentition is of the 

 utmost importance in investigations ot this 

 sort. The milk dentition, as we have already 

 said, is the family treasure which preserves 

 the characters by means of which the front 

 teeth, including the premolars, were marked 



in the ancestral forms. The true molars are 

 afterwards added to the milk dentition; but 

 they also existed in the ancestor, and for the 

 most part, indeed, in considerable number. 

 Now we ha\'e seen that no carni\ore now 

 living has more than one carnassial in each 

 half of each jaw, and this tooth, so highly 

 characteristic of the order, is always the last 

 premolar in the upper jaw, and the first 

 molar of the permanent dentition in the 

 lower. In all genera which have well-de- 

 veloped carnassials these exist, even in the 

 milk dentition ; but then it is always the second 

 last premolar in the upper jaw and the last 

 in the lower which exhibit this characteristic 

 form. The carnassial accordingly moves 

 further back at the shedding of the teeth. 

 This is particularly obvious in the hyaenas, in 

 which the carnassial of the milk dentition is re- 

 placed by a thick conical tooth in the permanent 

 one. From these facts it may, in my opinion, 

 be inferred that the ancestors of these carni- 

 vores must have possessed two more or less 

 developed carnassials, the first of which has 

 been lost in the permanent dentition which 

 now characterizes their descendants. Two 

 such carnassials, more or less fully developed, 

 are found, on the one hand, in many of the 

 marsupials of the present day, as, for example, 

 in the Tasmanian devil (Dasyurus), and, on 

 the other hand, in many extinct types, like 

 Hyaenodon, Pterodon, Cynodon, &c. We 

 may accordingly assume that these old types 

 are connected with the canmivores of the 

 present day by direct generation. 



We have thus summed up the facts. They 

 show, as it appears to us, that the primitive 

 stocks of the Carnivora must be sought in the 

 Marsupialia, which existed long before the 

 Tertiary period, and that these stocks were 

 from the first numerous. It is highly 

 probable that certain families are directly 

 descended from marsupial ancestors. The 

 Viverrida are beyond question in this 

 position. Their dentition approaches so 

 closelv that of the insectivorous and carni- 



