' 



■ 



204 Journal of the Asiatic Society of Bengal. [June, 1914. 



V 



even the dominant form in Ceylon. Moreover, although one 

 symmetrical genus of Aeeraiinae — Tiberioides — is found in the 

 northern parts of the area dominated by the genus Aceraius, 

 a comparison of the structure of the upper surface of its 

 head with that found in the other genera of the subfamily, 

 seems to show that this genus is a lateral offshoot from the 

 main trend of evolution, and is related to the other genera only 

 through the one symmetrical species found in Ceylon — Epis- 

 phenus moorei — which must therefore be regarded as the most 

 primitive existing species of the subfamily. 



From this it appears that the species of Aeeraiinae inhabit- 

 ing Ceylon are less highly specialized than those inhabiting the 

 Indian Peninsula; and that those inhabiting the Indian Pen- 

 insula are less highly specialized than are those found on the 

 other side of the Ganges taking these as a whole 



In the Gnaphalocneminae the three most primitive genera 

 are confined to Australia, except for one species (Episphenoides 

 pectinigera. Heller) from New Guinea, tlie remaining genera 

 being distributed over the East Indian Archipelago and Malay 

 Peninsula, one species penetrating into Bui ma as far as Tavoy. 



The line of demarcation between the Oriental and Austra- 

 lian Regions separates, almost completely, the Aeeraiinae and 

 the Gnaphalocnemis group of the Gnaphalocneminae from the 

 other groups of the latter subfamily, the only transgressors of 

 the line yet recorded b ing two sp /cies of the genus Gonatas, 

 and one of the genus Gnaphalocnemis. 



In order to explain the geographical separation of the 

 primitive symmetrical and closely related forms found in the 

 two regions, by the more highly specialized and less closely 

 related allies of ea^h, it must be supposed that conditions on 

 either side of ; ' Wallace's Line" are for some reason peculiarly 

 favourable to the evolution of highly specialized forms ; and 

 that these have migrated outwards, driving before them trie 

 less highly specialized, which have rarely survived to the pre- 

 sent day except where they have been able to establish them- 

 selves behind zoogeographieal barriers that the more recently 

 evolved forms have not yet been able to cross. 



This is further supported by the fact that of the less highly 

 asymmetrical forms which are not un< ommou in the Australian 

 half of the East Indian Archipelago, one whole group at least 

 (comprising the genera H y per pb sthenics, Labienus, Kaupiolus 

 and Aurehus) shows a high degree of specialization, in struc- 

 tures which are perfectly normal in most of th more highly 

 asymmetrical forms among which they live. - 



There is a curious similarity between the relation ot 

 specialization to geographical distribution found in the asym- 

 metrically inclined Passahdae, in the Thelyphonidae (see 

 Gravely, J.AS.B., VII [8], 1911, Proceedings, pp. exxiii-exxv) 

 and in the Crinoidea ^see Clark, Echinoderma of the Indian 



