50 ME. A. MUBBA-T ON THE GEOaBAPHIOAI- llELATIONS Or 



US of reliable data for estimatiBg the affinity of the inhabitants of 

 different regions which we should otherwise have had. 



I give a list of the known species contributed from other coun- 

 tries in the Appendix, from which it will be seen that they are 

 derived as follows, viz. from 



North America ^.... 2 timber species (common) . 



Brazil 3 timber species (common). 



, f 4 timber species (2 micro- 

 East Indies and Philippine Islands | ^^^^^ ^^^ ^ carabid). 



r2 timber species, 2 Hetero- 

 New Holland ^ mera, both microtypal, and 



1^ 1 Staphylinid, microtypal . 



f 1 timber species and 1 Sta- 

 New Zealand... | ptyUnid, microtypal. 



And 1 cosmopolitan wanderer of doubtful origin. 



Besides these, however, there are a number of types from these 

 countries which have probably been introduced by nature without 

 the help of man ; for they are modifications of peculiar forms, and 

 not actually the same ; and whatever may be the alterations which 

 change of climate has made within the reach of man's observations 

 on species of the higher animals introduced by him, there is not 

 the slightest ground for supposing that any change has been ever 

 so produced in insects. It was, of course, only to be expected 

 that there should be something of this sort from the East Indies. 

 It is natural that there should be some overflow or some dispersal 

 into a country so near ; and in conformity with this we find the 

 proportion of its overflowings diminishing as we recede from 

 India. It is for this reason that I have in my Table above men- 

 tioned limited the list to Eairmaire's species from Central and 

 Eastern Polynesia. Even that is divisible into two sections. Of 

 the Eastern, M. Pairmaire says : — 



" At Tahiti, where the temperature is very various, owing to 

 the lofty mountains, where numerous streams preserve freshness, 

 the species are more varied, the individuals in greater abundance, 

 Carabid and Braclielytra are met with. The Sandwich Islands, 

 which are situated to the north of the Equator, at the same 

 distance as Tahiti is to the south, give almost the same insects 

 in very small quantity." — Bev. Zool. 1849, 279. 



Of the middle portion, again, ho says, " At Tongatabou, a flat 



