pri;u(»i'us. S;^ 



■R'liereas I't. tuDnpi/rtis is an iiiclo-Mulayau type wliicli has spread 

 only to the south-west corner of the ])re.sent area. 



Also the negative features of the Austro-Malayan Pteropys fauna 

 must be referred to here. The five jjroups of the genus which 

 have no representatives in Austro-Malaya are the Pt. mnriannus, 

 samo'ensis, ^'seJaphon, vieJanotus, and i-ufus groups. The two 

 former are entirely Polynesian, but one of them, the mariannus 

 group, is phylogenetically very intimately connected with the 

 Austro-Malayan /t//y;o*»eZa/u<« group, and the other is closely related 

 to the Anstro-Malayan Pt. lomhocensk group. The pseiaphon 

 group is Micronesian and Philippine, but as lieing closely connected 

 with the samoetuiis grou]) leads again back to Pt. lomhocensis. The 

 melcmotus and rafua groups are geminate types, the one confined 

 to the Audaman-Nicobar chain, the other to the Malagasj' region, 

 and related to the hjipomelanus group. 



The two facts, that the Pteropus fauna of Austro-Malaya is incom- 

 ])arably richer than that of any other zoogeographical area of similar 

 size, and that all groups of the genus not represented in Austro- 

 Malaya are very closelj' related to Austro-Malayan types, /. e. 

 pointing back in their probable origin to Austro-Malaya, lend 

 support to the hypothesis that the centre of dispersal of the genus 

 was this subregion, or rather an area much larger than, but in- 

 cluding, Austro-Malaya and probably excluding Australia. But 

 any hypothesis is unsafe, so long as we know very little about the 

 physical changes that have taken place in the area between 

 the Malagasy region and the Indian Archipelago, and nothing 

 about the past history of the genus. 



V. Australia. — 5 species, of five groups, viz. : — I't. coiiKpiciUaius 

 (Pi. con.tpicillutu.t group), Pt. poUocephalun (Pt. mairuiis group), 

 Pt. goiiJdi (Pt. a/ecto group), Pt. brnnneus {Pt. Iii/poDirhiniis group), 

 Pt. srapulains {Pt. xra/ndntits group). 



The small number of species, representing as many groups, is 

 evidence that the genus forms no part of the original fauna of the 

 continent. The first species, Pt. consjudllatus, is common to S.E. 

 New Guinea and Australia and has very close relatives in W. Xew 

 Guinea and the Moluccas ; the second, Pt. poVioci'jthalnx, is a j)eculiar 

 modification of a New Guinean type. The third species, Pf.^jouhii, 

 points toward the Lesser Sunda Islands and Celebes, being very 

 closely allied to J't. alccto (Celebes, Sahn'er, Lombok). The fourth 

 and fifth species, Pt. hrunnens and srapulotus, have their closest 

 relatives in the Solomon Islands. Of the five Australian species, 

 I't. •^coptilatns and poliocepJuihis are decidedly the mo*;t peculiarly 

 modified, hence perhaps the earliest immigrants into the continent. 



VI. Polynesia. — 18 species, representing live groups, viz. : — (1) 

 ft. pdewensis, ijapenxis, tiaJaims, marmnnnx, vdiiih-orensifi, f/eddii-i, 

 tiiiii/anus {Pt. ituirlaiinus group) ; (2) /''/. lutwmeiisis. namoi'iisi't, 

 unetJanvs {Pt. R(tmo'n\s'is grou])) : (3) Pt. InsuJans, filiaocephalus , 

 jisfliiphon, pilosKs. ttiJi.txi'latiis {Pt. pselaphon group); (4) Pt. 



G 2 



