EVOLUTION 403 



correlation determine the numerical specification of each 

 form of life, and when we say that evolution is taking 

 place, we mean that progressive changes are going on in 

 one or all of the numerical values which fix the mean, 

 variability, and correlation of the system of organs and 

 characters. Within the organism, if it puts forth like 

 organs — flowers, seed capsules, leaves, blood corpuscles, 

 hairs, scales, etc., which are in whole or part un- 

 differentiated ^ — we may have an individual type, an 

 individual variability, and an individual correlation." 

 These must be carefully distinguished from the racial 

 type, racial variability, and racial correlation. It is 

 indeed this distinction which makes the organism an 

 individual ; it grows its like parts more alike to each other 

 than to those of another individual ; it grows its unlike 

 parts more closely associated with each other than with 

 those of another individual. If a hand or a whole skeleton 

 were put together at random, we should be able even with 

 our present knowledge of correlation to give some idea 

 of the odds against such an individual having ever 

 existed. If fifty beech leaves were picked up at random, 

 we could obtain some appreciation of the probability that 

 they had grown on one tree. A very important question 

 then arises, namely : Do all individuals of whatever race 

 grow in the same manner ? Is the correlation between 

 pairs of undifferentiated like organs in the individual the 

 same or nearly the same for all forms of life? If so, we 

 have ascertained quantitatively as comprehensive a law of 

 growth for living organisms as the law of gravitation for 

 molar masses. My researches on this point are not yet 

 complete, but they indicate that the following law is true. 

 The degree of resemblance between undifferentiated like 

 organs in the individual is nearly the same for all forms 



1 The word undifferentiated must be emphasised. Two fingers are like 

 organs, but they are differentiated in both size and function. Two leaves can 

 also be differentiated, as may frequently be observed in leaves near the flower 

 or fruit, e.g. in ivy or in Spanish chestnut trees, etc. 



^ E.g. we might find the correlation between weight and veining on leaves 

 of the same tree, or between length and breadth of blood corpuscles in the 

 same frog or newt, etc. 



