1 30 RETROGRESSION 



understand it. Presumably a * contribution ' to the heritage of a 

 descendant means more than mere resemblance; otherwise it is 

 difficult to understand why a word so apt to mislead is used when 

 the unequivocal word ' resemblance,' which occurs readily to the 

 mind, is sufficient to describe the phenomenon. An ancestral con- 

 tribution which is halved and quartered and so forth would 

 certainly seem to imply a discrete unit representing an ancestor in 

 the germ-plasm ; and apparently since every ancestor's heritage is 

 compounded from those of his predecessors, it would appear also 

 that the contribution of every ancestor is represented in the 

 contribution of every successor. To say the least, it is hard 

 to believe that machinery for this perpetual and complicated 

 division can exist. 1 



of the ancestry ', may be termed the Law of Ancestral Heredity " (Mr Udney Yule, 

 " Mendel's Laws and their Probable Relations to Intra-racial Heredity," The 

 New Philologist, 1902, p. 202. " The degree to which a parental character affects 

 offspring depends not only upon its development in the individual parent, but on 

 its degree of development in the ancestors of that parent " (Prof. W. F. R. Wei don). 

 Mr Yule and Professor Weldon avoid the use of the word contribution. As 

 formulated by them the Law expresses what would appear an obvious truth ; 

 but it is then widely different from the Law as formulated by Galton and 

 accepted by Mr Vernon. 



1 I have reasoned from what appears to me the plain meaning of Gal ton's 

 words. Here is another and apparently very similar interpretation : "A man 

 may receive a quarter of his hereditary characters from each parent, and a six- 

 teenth from each grandparent, but all except a] very minute portion of these 

 characters are common to all men, they being, in fact, the characters proper to 

 the species, Homo sapiens, as such. Instead of a quarter of a unit from each parent 

 a man receives only a hundredth or a thousandth of a unit of characters peculiar 

 to the parents as such, all the rest being the characters common to all the members 

 of the race. Even this minute fraction of a unit does not in any way represent 

 characters acquired by the parent during his lifetime, but is itself built up of 

 proportions of peculiar characters received from his parents, grandparents, and 

 other ancestors in accordance with the law of heredity " (Vernon, Variations in 

 Animals and Plants, pp. 146-7). Presumably Dr Vernon does not intend to imply 

 that a man receives a quarter of his bulk, his mass, from each parent. Extension 

 and weight are derived wholly from nutriment. We must go back, therefore, 

 as always in the study of heredity, to the germ-plasm, or rather to its hereditary 

 potentialities. If I understand Dr Vernon aright, he accepts the theory of discrete 

 ancestral units which are halved and quartered and so on in the germ-plasm, but 

 thinks that all ancestral units, at any rate all units that have come into existence 

 since man evolved, are much alike i.e. he thinks that, apart from the potency 

 due to bulk, units differ only in comparatively^insignificant details ; just as men 

 are alike in being men, but differ inasmuch as they vary somewhat from one another. 

 The hand is a character " proper to the species, Homo Sapiens, as such." Is it 

 then contended that the child derives one-quarter of his tendency to develop 

 that organ under the stimuli of nutrition and use from each parent, one-sixteenth 

 from each grandparent, and so on ? The theory of Ancestral Inheritance assumes 

 that the influence of an ancestor eleven generations back is on the average more 

 than a billion times weaker than that of parent. But the character of a particular, 



