4i 8 H. M. KYLE [December 



detail of description and fact without law or connection, or tending too 

 much the other way, show too much of the observer and become guilty 

 of anthropomorphism. The latter attitude is but too prevalent amongst 

 opponents and upholders alike of the theory of natural selection. 



The attitude here advocated lies between these two. The details 

 and facts are welcomed, but " facts " do not make true knowledge, and 

 on the other hand, any theory which makes the observer inclined to 

 read into the facts his own personal notions of " utility " or " advantage," 

 for example, must be considered as too prone to misinterpret the actual 

 phenomena. What we wish rather is the attitude of a philosopher 

 who perceives well the facts, but holds himself aloof from opinion, and 

 seeks some method as an intermediary and aid to interpretation. What 

 this method should be is not difficult to conceive. 



The outstanding feature in the Darwinian hypothesis — seized upon 

 by its clerical critics at the first appearance of the Origin of Species — 

 was the stress laid upon " chance." The conception therein involved 

 was — not that anything ever happened at random or haphazard, but 

 that the changes occurring might be conveniently so expressed. As is 

 well known, this theory of chance has been developed more and more 

 of recent years as a separate study, until the biologist has come to 

 regard it with suspicion as something foreign to his own subject. And 

 yet is there not some truth in this theory of chance ? And again, have 

 natural selectionists the prerogative of this truth ? 



The truth underlying this theory is not far to seek. Life, we may 

 say, depends on many "chances." Hence, knowing the various sources 

 of danger, we may, as if insurance agents, calculate the " chance " of a 

 particular individual surviving to a certain age. In a deeper sense, 

 again, if we knew the causes of variation we should be able to calculate 

 the " chance " of the appearance of any particular variation under 

 certain conditions. What we have before us at any time is only 

 partially the truth, and even if we knew all that had ever occurred and 

 understood all, we — not being omnipotent — could only state as a 

 probability what would occur next. 



This is the justification for the theory of " chance," or we should 

 rather say of " probability." 1 Has the theory of natural selection 

 any exclusive right to this conception ? 



We may judge of this more closely by following the facts grouped 

 under the three propositions stated above. From (1) we learn that 

 the variations in any character are naturally grouped about a certain 

 average, and are usually distinguished as plus and minus variations from 

 that average. If then we hold by natural selection, and maintain that 

 hy this law evolution may proceed by slow minute steps, so that "even 

 a grain in the balance shall decide which shall live and which shall 

 die," — which variations shall we call " useful " or " favourable " ? If it 



1 A mathematical description of the meaning underlying these words will be found in the 

 "Chances of Death," by Karl Pearson. 



