Natural Selection still Necessary 211 



There are positive grounds, indeed, to take the matter 

 further, for discarding the extreme position of those who 

 deny the part in evolution played by congenital variations 

 accumulated by natural selection. The specific character, 

 the persistence, and the definiteness of the hereditary im- 

 pulse, require that we should recognize its leading role in 

 development, despite the large part attributed to individual 

 accommodation. All the evidence accumulated by writers 

 since Darwin in support of natural selection operating 

 upon variations, together with the statistical work upon 

 variations, is available to show that heredity represents 

 a real and definite impulse which conserves the specific 

 type, and in large measure the specific characters, of 

 organisms. Recent work in experimental morphology 

 emphasizes the persistence of heredity in reverting back 

 to its normal development in the individual, as soon as 

 artificial conditions under which it may have been modified 

 are removed. 1 This persistence appears in the life history 

 of twins, in the phenomena of atavism, in ' exclusive ' as 

 opposed to 'blended' heredity, 2 in the protected develop- 

 ment of pupae, etc. 



This granted, variations in congenital endowment at 

 once become liable to the operation of natural selection for 

 any utility they may serve, and this the more when they 

 are supplemented by individual accommodations. 



An additional general remark is suggested by this criti- 

 cism. It should always be borne in mind that a theory 

 of evolution does not attempt to account for organs nor 



1 This has been dwelt upon by Professor E. B. Wilson. 



2 On the persistence of hereditary traits in twins see Galton, Enquiries into 

 Hwnan Faculty, pp. 216 ff. On 'exclusive' Heredity (i.e., heredity from 

 one parent only) see Galton, Natural Inheritance, p. 12 (cf. also the Index to 

 that work). 



