56 THE VARIATION OF ANIMALS IN NATURE 



unconvincing. At the same time we do not pretend that the 

 evidence available suggests that any ' Lamarckian ' process is 

 very important as a source of new heritable variation, except 

 possibly in the matter of habits. There is certainly a very large 

 body of evidence (Chapters IV and VII) suggesting that the 

 bulk of the morphological differences between species and races 

 is not in any way correlated with a particular environment ; 

 and conversely that many species and subspecies range widely 

 without any modification. Although this seems in conflict 

 with the evidence for geographical trends (p. 46), yet such 

 trends are relatively uncommon (i.e. compared with the 

 number of races and species not arranged in trends) and 

 further usually only some of the characters of a species exhibit 

 the trend. 



Conclusions. 



In this chapter we have considered the origin of the various 

 types of variation that may be encountered in a natural popu- 

 lation. Fluctuations certainly form a large element, but 

 quantitative data as to the importance of these are hard to 

 obtain. Genetically determined variations include (in addition 

 to gene-mutations) changes due to fragmentation, etc., of 

 chromosomes, polyploidy and recombination. The first two 

 phenomena seem to be of minor importance in animals. Re- 

 combination is certainly responsible for much of the normally 

 wide range in phenotypes. We have not much evidence 

 yet whether species in nature are often heterozygous for 

 more than a few characters. If they are not, the results of 

 recombination are strictly limited, especially in any particular 

 direction. In any case Lotsy's theory of evolution by crossing 

 cannot have much application in the animal kingdom, where 

 successful interspecific crosses are relatively uncommon. 

 Gene-mutations are certainly a very important source (or, as 

 some would have it, the only source) of new hereditary material. 

 The real cause of gene-mutations is quite unknown, but it is 

 theoretically improbable that they are in any real sense 

 spontaneous. The rate at which they occur has now been 

 influenced by X-rays, radium-rays and high temperature. 

 Even under these influences the rate is still relatively low. 



The problem of the inheritance of induced modifications 

 appears to be ultimately reducible to the question whether 



