THE SYSTEMATISTS 505 



understand the reasoning of the mathematician or astronomer 



without a prolonged preliminary training in thinking. The mental 



attitudes of the followers of the two kinds of sciences are markedly 



different. Thus, I believe every student of the interpretative 



sciences would regard it as better science if the theory that 



the individual recapitulates the life-history of the race were first 



inferred as a necessary consequence from the familiar truth that 



\ offspring recapitulate with variations the development of their 



'I parents, and then proved by an appeal to the less familiar facts 



jj of embryology; whereas I think some biologists would regard 



1 such a proceeding as unscientific, and would prefer to begin at 



| the other end and found an untested hypothesis on the evidence 



| furnished by embryology alone. Biology is remarkable for the 



number of untested and often untestable hypotheses which its 



followers have formulated. Witness the numerous hypotheses 



concerning inheritance, sex, immunity, and cancer. On the other 



hand, if a physicist formulates a hypothesis he is almost sure to 



endeavour to furnish not only the evidence on which it is founded, 



but also that by which he thinks it is proved. Witness Kepler's 



rejection of nineteen of his own hypotheses, which he disproved 



by a deductive appeal to reality, before he arrived at the true 



statement of planetary motion, and Newton's suppression for 



fifteen years of a true theory of the moon's motion, because it 



did not accord with data which were subsequently found to 



be wrong. 1 



825. Obviously, since the mathematician has only a few data 

 on which to found his thinking, his science must be deductive 

 from its very beginnings. On the other hand, the biologist, like 

 the astronomer, must begin by classifying his facts according to 

 co-existences and resemblances, and only afterwards, when they 

 have thus been rendered manageable, proceed to construct chains 

 of causation. At any rate, every science, the facts of which are 

 very numerous and which has passed beyond the descriptive 

 stage, has grown in this way. 2 



826. Darwin attempted to make this second step. In his theory 

 of Natural Selection he formulated that which, if true, is a real 

 law. He described a uniformity in the sequence of events from 

 which necessary consequences could be deduced. He strove to 

 interpret facts about living beings, and so to link them together 

 in chains of causation. Consciously or unconsciously he sought 

 to do for biology what Newton had already done for physics and 



1 See 72. 2 See 69 (footnote). 



