474 MR. O. THOMAS ON THE MAMMALS [DeC. 4, 



To this view 1 have come chiefly by finding that such rudiments of 

 the distinguishing characteristics of Pteralopex as are present in other 

 Pteropodidse are not confined to one or two species of Pteropus, but 

 are found scattered about in the different genera of the family, as 

 though they liad been independently inherited from common ancestors. 

 Thus, while Pteropus leucopterus, in addition to its completed orbit, 

 resembles Pteralopex in having basal ledges on the posterior sides of its 

 incisors, and Pteropus aneiteanus in having its cheek-teeth so grooved 

 transversely as well as longitudinally as to recall those of the new 

 genus, yet Cynopterus has frequently bicuspid upper canines ; and, 

 above all, the nearest and most significant resemblance is presented 

 by the otherwise very different genus Harpyia. There, not only do 

 the upper canines have a distinct postero-external secondary cusp of 

 the same relative development as that found in Pteralopex, but the 

 three chief lower cheek-teeth, i. e. the two posterior premolars and 

 the anterior molars, have absolutely the same primitive " tuberculo- 

 sectorial" form as those oi Pteralopex, the individual cusps homo- 

 logizing perfectly with those of that animal. In addition, the upper 

 cheek-teeth of Harpyia present something of the same primitive 

 character ; and therefore, judging merely by dentition, that genus 

 should be looked upon as being in a still more generalized state than 

 even Pteralopex. 



Should this view of the origin of the dental characters o{ Pteralopex 

 be even approximately correct, it is clear that the reputed relationship 

 of the Macrofflossi with the Olossophagce must have no real founda- 

 tion in fact, since the common ancestors of the Macro- and Micro- 

 chiroptera having had cuspidate teeth, and, no doubt, insectivorous 

 habits, these groups, whose likeness lies in their small non-cuspidate 

 teeth and frugivorous habits, cannot be little modified descendants of 

 the Palseochiroptera, but must be independent and comparatively 

 recent offshoots from the two great groups to which they respectively 

 belong. 



I imagine, then, the history of the evolution of the present groups 

 of Chiroptera to have been somewhat as follows' . The earliest Bats, 

 or Palseochiroptera, would have been cuspidate-toothed and insecti- 

 vorous like their ancestors the terrestrial Insectivora. Among them 

 there would presently have arisen a form like Harpyia^, fruit-eating, 

 but still with cuspidate teeth and no doubt markedly " tuberculo- 

 sectorial" premolars and molars. Then, while the modern Harpyia 

 would have arisen in one direction by the reduction of the incisors, 

 in another there would have followed some form like Pteralopex, 

 still retaining to a certain extent cuspidate teeth. Then the cusps 

 ■would have more and more tended to disappear, the result being 

 Pteropus and its allied genus, of which some few species (e. g. Pteropus 



' Compare also Prof. W. Leche'e learned and philosophical remarks on the 

 pbylogeny of Galeopitkecus, an animal which, according to him, is a much 

 modified representative of the ancestors of tlie Chiroptera at a time when they 

 were, so to speak, just learning to fly. (K. Vet. Ak. Handl. xi. no. 11, 1886.) 



^ Of course this ancestral Harpyia would have had well-developed incisors 

 above and below, as iu the majority of PteropodidcB. 



