770 



REPOUT -1884. 



i ii 



6. On some Peculiarities in the Geographical Dislrihution of certain Mamimh 

 inhabiting Continental and Oceanic Islands.^ By G. E. Dobson, Jf i 

 F.B.S, 



Tbo author points out the interostinf^ fact, not hithiirto noticed, that many uf 

 the most characteristic species of the Chiropterous fauna of Au.straUa Imve tluir 

 nearest allies not iu the Oriental but in the Ethiopian region, thus contrastiiiL' 

 remarkably with the avi-fauna, and instances the presence of species of the geiima 

 Chaliiiolohus and MurinoptcruH as occurrinp; in tlie Ethiopian and Australian repoiij, 

 while they are unrepresented in the Oriental ; and also the stroii}^ natural atHniti.< 

 existing between certain species of other gent 'ra inliabiting the former n'gions. I!,. 

 also draws attention to tlie fact that not less than 80 per cent, of the species uf 

 Pteropus " are restricted to Australia ainl Madagascar with its islands, whil.' a 

 single species only, evidently derivativi', inluibits the immense continent of India 

 and the island of Ceylon. The necessity, in the first place, of postulating sume 

 connection between Africa, Madagascar, and the Australian Continent, to 

 account for tlieir distributional peculiarities, appears evident from oonPidfratioD 

 of the above noted facts, but the writer remarks that such connection was 

 not by way of India, as shown by the absence of connecting spe-.'ies in 

 that country, and points out that we are therefore obliged to suppose tliiU at a 

 comparatively recent period a chain of islands connected these continents, the 

 islands being sufBciently far apart to prevent the entrance of terrestrial mammal-, 

 yet near enough to permit of the occasional passage of some of the flying species. As 

 it might with reason.be urged that such pi'opinquity of islands would also permit of 

 interchange of the avi-fauna, the writer remarks that the existence of a complete 

 chain of islands separated by sulliciently narrow straits may have existed for a 

 short period only, the completeness of the chain being, perhaps, dependent on 

 some volcanic group, which may have disappeared as suddenly as it came into 

 existence; under such circumstances bats would be much more likely to estabM 

 themselves successfully iu the new contiuc^ntal lands open to their migrations 

 for the following reasons: — (1) that tlie food of both the frugivoroiis and in- 

 sectivorous species is of a more general character than that of birds, few of the 

 species of w-hich are so omnivorous, within these limits, as the bats ; ('2) tliat the 

 nocturnal habits of the bats would enable them to escape observation from enemies 

 always sure to recognise the presence of solitary individuals. 



With respect to the peculiardistribution of the species of the genus P/wo/jh-«, includ- 

 ing the great frugivoro us bats, it is a noticeable fact that, although the small islands 

 of Mauritius, Bourbon, and the Comoro group have each two very distinct species, 

 the great continent of India and Burma and the island of Ceylon have but one. It 

 appears, thei^efore, probable that India owes its single flying-fox to soma otlier 

 region, and in seeking for the country from which it is derived we must consider 

 its nearest allies among the species of the genus. 



Now this species differs from P. Edwardm of Madagascar and the Seyclielle 

 Islands in few and unimportant characters, presenting such differences only as 

 might have resulted in a few generations, though they now appear to be permanent- 

 It is, therefore, evident that these two species have been derived within a compara- 

 tively recent period from a common ancestor, and probable that the enormous 

 number of individuals of P. medius, now representing the genus in India, are the 

 descendants of a few individuals originally escaped from their island homes in the 

 Indian Ocean, and now cut oflf by subsidence of some of these islands from their 

 nearest relations. On no other hypothesis can we account for the discontinuous 

 distribution of the species of this genus, for the incapability of these large bats to 

 traverse wide oceanic spaces is shown by the fact that the narrow channel of 

 Mozambique between the Comoro Islands and the African coast has sufficed to 

 prevent their entrance into Africa, where tree fruit is abundant, and where immense 



' This paper will be published in extenso in the Annals and Magazine of Natural 

 Bistort/, series 5, vol. xiv. 

 ' Flying-foxes. 



'^■-'- ' -^" 



