DARWIN'S WORK 213 



of Natural Selection. His reasoning was largely deductive, for, apart 

 from the tests to which he subjected his thinking, his attention was 

 not attracted by human disease and, consequently, he never observed 

 Natural Selection actually at work. He was unable, therefore, to 

 reach his wider synthesis through an actual observation of parti- 

 cular facts or test it by an appeal to them, but his work was fertile 

 in suggestions hence, for example, Poulton's attempt to demon- 

 strate Natural Selection amongst chrysalises, and Weldon's 

 attempt to prove its occurrence amongst crabs. Hence also many 

 biometric inquiries. 



352. Starting from the point at which Darwin stopped, accept- 

 ing his theory of evolution by Natural Selection, and combining 

 with it the hypotheses which we reach by induction, that species 

 tend to retrogress on cessation of selection, and that offspring 

 recapitulate the main features of the parental development, we are 

 able to infer (i) that retrogressive variations tend to predominate 

 over progressive variations, (2) that, with rare exceptions, variations 

 are spontaneous, and that spontaneous variations tend to occur all 

 round the specific mean, (3) that, therefore, the germ-plasm is 

 highly insusceptible to change through the direct action of the en- 

 vironment, (4) that, apart from his own progressive variations, the 

 development of the individual is an abbreviated and inaccurate 

 recapitulation of the evolution of the race, and (5) that, therefore, 

 every retrogressive variation is, in effect, a reversion. Here we 

 formulate laws or generalizations of heredity which are probably 

 as nearly fundamental as it is possible to reach in the present state 

 of our knowledge. As in the case of Natural Selection, the process 

 of reasoning, since we endeavour to ascertain whether our laws are 

 in harmony with one another and with reality, is largely deductive. 

 But, granting the premises from which we started, and none of 

 them are seriously disputed, I believe we must not think at all, or 

 we must think in terms of miracle, or we must accept the laws as 

 true. And, very obviously, if they are true, if they are real laws, 

 we have made a step towards rendering biology (or at least the 

 study of heredity and evolution) a deductive science towards 

 establishing it on the basis of a few generalizations, to which a vast 

 mass of data, already discovered or yet to be discovered, may be 

 linked by chains of causation, that is to say, towards attaining 

 that goal which is the aim of all science. 1 



1 " It will be seen hereafter that there are weighty scientific reasons for giving 

 to every science as much of the character of a Deductive Science as possible ; 

 for endeavouring to construct the science from the fewest and the simplest possible 



