416 OUTLINES OF E VOLUTION ABY BIOLOGY 



fluctuating variations under the control of natural selection as 

 that they originate in any other way that can be suggested in 

 the present state of our knowledge ; and even if these variations 

 are at first purely somatogenic, we may suppose that in the 

 course of many generations they gradually exert a cumulative 

 influence upon the germ plasm until the latter, so to speak, 

 topples over into some new position of equilibrium and a new 

 unit character arises. We cannot, however, now add to what we 

 have already said on this subject when dealing with the theory 

 of heredity. 



One of the most important factors in bringing about divergent 

 evolution is undoubtedly isolation. However a new character 

 may have arisen it is liable to be swamped by the crossing of 

 the individuals which possess it with others which do not possess 

 it unless by some means or other the two groups are prevented 

 from interbreeding. We cannot fail to see the importance of this 

 principle when we study the fauna and flora of remote islands, 

 and their relationships to those of the nearest continental areas 

 or of other islands. 



Take, for example, the case of the Chatham Islands, which, as we 

 have already seen, lie some 400 miles to the east of New Zealand. 

 There can be very little doubt that these islands were formerly 

 connected with the New Zealand mainland, and this connection 

 probably continued into Pleistocene times, when a great 

 depression took place which caused the two to be separated by 

 a wide tract of ocean. All who have studied the question are 

 agreed that the fauna and flora of the Chatham Islands are 

 simply isolated detachments of those of New Zealand. Many 

 species, especially of the plants, are identical with New Zealand 

 species, but many others, though closely related to those of New 

 Zealand, are considered by systematists to be specifically distinct, 

 and they occur nowhere else in the world. 



We have here an excellent illustration of the effects of 

 geographical isolation, which we shall be able to appreciate 

 better, perhaps, if we confine our attention to a few typical cases. 

 The common New Zealand wood pigeon, Carpopliaya (Hcmiphagd) 

 novce-zealandife, is represented on the Chathams by a species 

 known as Carpophaga (Hemiphaga) chatliamensis, differing but 

 slightly from its new genus congener, and the New Zealand 



