1914-] F- H. GravEIvY : An Account of the Oriental Passalidae. 315 



The general nature of the evolution and distribution of this subfamily, together 

 with that of the Aceraiinae, is shown in the accompanying figure (text-fig. 7). From 

 this it will be seen that the forms of Gnaphalocneminae which come nearest to the 

 most primitive of the Aceraiinae ^ — Episphenus moorei from Ceylon — are Australian ; 

 so it seems that in the Gnaphalocneminae also, the most primitive forms have been 

 replaced by more highly specialized ones in the central parts of the Indo-Australian 

 area, but have survived where they are cut off from aggression by some geographical 

 barrier which the latter have as yet been unable to cross. 



The only simple and symmetrical forms of Gnaphalocneminae found outside 

 Australia are : (i) a single species — pectinigera, Heller — from New Guinea, of the other- 

 wise purely Australian genus Episphenoides; and, (2) a few species of the closely allied 

 genera Cetejus and Analaches. All the remaining extra- Australian symmetrical species 

 are specialized in other ways. For instance, in the whole of the H y perplesthemis group 

 of genera, no member of which is very highly asymmetrical, and some of which are 

 perfectly symmetrical, the lateral and intermediate areas of the metasternum are fused ; 

 and in addition, the central tubercle of Aurelius, and the antennae of Labienus, are of 

 somewhat unusual form. 



A comparsion of the geographical distribution of the Aceraiinae and Gnaphalo- 

 cneminae with that of the Arachnid family Thelyphonidae, is not without interest. 

 The Passalidae and Thelyphonidae (of the Oriental Region at least) inhabit the damp 

 jungles of more or less hilly country ; both families are absent from the Gangetic Plain ; 

 and in India the northwestern boundary of the geographical range of both is approxi- 

 mately the same, being dependent in both cases probably on climatic conditions. 

 Finally, in the Thelyphonidae, as in the Aceraiinae and Gnaphalocneminae, we have a 

 structural index to the degree of specialization found in different species, and find that 

 species are most numerous and most highly specialized in the neighbourhood of Borneo 

 and Celebes, and less numerous and less highly specialized the further one goes from 

 these islands in any direction.^ This seems to indicate a radial pressure of highly 

 specialized on less specialized forms, resulting in a radial migration, one line of which 

 especially — that in a westerly direction — follows very much the same course as 

 appears to have been followed by the advance of more and more highly specialized 

 forms of the Passalidae now under consideration. 



The fact that this type of distribution is found in two such widely separated 

 groups of Arthropods suggests that it may occur in a number of other groups also. 

 And the peculiar richness of the fauna of the Malay Archipelago, which has long been 

 known, and has contributed largely to the advancement of the hypothesis that insular 

 conditions favour the rapid evolution of species, is evidence in favour of this sug- 

 gestion. But when, as appears to be the case in the beetles now under discussion, 



' Episphenus moorei appears tobe more primitive than any species of the genus Tiberioides. For the 

 latter, although equally symmetrical, differ from all other species of the subfamily in having the frontal 

 tubercles less widely separated, and so probably form a divergent line of descent (see diagram). 



^ See J.A.S.B. (N.S.) VII (1911), Proceedings, pp. cxxiii-cxxv. 



