﻿IO DARBISHIRE, Mendelian and Biometric Theories. 



cells. It is perhaps not unnatural, though it is certainly 

 unjustifiable, that, when an experimenter is thinking- of a 

 set of facts before him now in terms of the one theory and 

 now in terms of the other without having clearly fixed the 

 peculiar characters of each theory to its proper owner, he 

 should get these characters misplaced, that he should 

 add to the real character of the biometric theory one 

 which it does not possess — that of applicability to the 

 individual. I have discussed this particular error of judg- 

 ment at some length because it illustrates the kind of 

 mistake a student of heredity at the present time may 

 make unless he realises the exact nature, at which I have 

 so far only hinted, of the two theories whose compatibility 

 with fact he is testing. Were I not persuaded that mine 

 is not the only case in which this or a similar kind of error 

 has been made I should not have described it : and it is 

 because I believe that the misunderstandings and argu- 

 ments at cross purposes, which have lately characterized 

 discussions on heredity, will be things of the past when 

 the relation between biometric and Mendelian theory is 

 clearly seen that I set forth the following considerations. 



III. 



From a point of view which commands a wide range 

 of our experience, our knowledge may be divided into two 

 distinct classes, according as we are dealing collectively 

 with a vast number of things — with a mass phenomenon ; 

 or with the individual units which make up that mass. 

 These two kinds of knowledge are radically different, and 

 are distinguished from one another by the same characters 

 as those which are peculiar to the two meanings of the 

 statement that a thing happens by chance ; for when we 

 make this statement we may either be referring to the 

 method by which I decided whether to write \ biometric ' 



