THE RAW MATERIALS OF EVOLUTION 109 



they seemed to him to partake too much of the 

 pathological. But we must not hurriedly dismiss 

 mutations like that of fowls with webbed feet and 

 no tails as obviously teratological, for most of 

 them may be matched in nature. If there were 

 only one specimen of a cross-bill, for instance, 

 would it not be regarded as a freak which could not 

 possibly survive in nature ? In the second place, 

 reacting as he was against a catastrophic view of 

 nature, and looking at things (as he said) through 

 Lyell's' eyes, Darwin naturally fought shy of 

 big sudden changes. Moreover, as he said to Asa 

 Gray : " There seems to me in almost every case 

 too much, too complex, and too beautiful adapta- 

 tion in every structure to believe in its sudden 

 production." Finally, he thought that a full- 

 fledged new character appearing suddenly would 

 be swamped by intercrossing.* 



The last difficulty, which is the only serious one, 

 has been removed, for it is characteristic of muta- 

 tions that, when they arrive, they come to stay, 

 unless they be eliminated as disadvantageous. In 



* It was characteristic of the Lyellian, or Uniformitarian school 

 of geologists to explain large results on the principle of slow suc- 

 cessive increments, accumulating for a very long time. 



2 Let us hear what he says in the last edition of " The Origin 

 of Species " : 



" Mr. Mivart is further inclined to believe, and some naturah'sts 

 agree with him, that new species manifest themselves ' with sudden- 

 ness and by modifications appearing at once. . . .' This con- 

 clusion, which implies great breaks or discontinuity in the series, 

 appears to me improbable in the highest degree " (p. 201). 



" Although very many species have almost certainly been 

 produced by steps not greater than those separating fine varieties, 

 yet it may be maintained that some have been developed in a 

 different and abrupt manner. Such an admission, however, 

 ought not to be made without strong evidence being assigned " 

 (p. 203). 



