FILIAL REGRESSLON 321 



of fathers on children, so that the fathers of abnormal children 

 are, on the whole, less abnormal than their children, they would 

 either have to attribute this feature of regression to a vital 

 property by which children are able to reduce the abnormality 

 of their parents, or else to recognise the real nature of the pheno- 

 menon they are trying to discuss. 



" It seems likely that in cases where the mating of parents 

 is not determined to any serious extent by their likeness or 

 unlikeness in the character discussed, the regression of children 

 on parents has a value very nearly the same, and very nearly 

 equal to \, for a large series of characters, mental as well as 

 physical, in human beings, and for a large series of characters 

 in the higher animals, at all events, if not in animals generally" 

 (Weldon, 1906, p. 108). 



Summary. — Many individual organisms do not differ much 

 from the mean of the race to which they belong. We cannot 

 say " from the species to which they belong," because many 

 systematic or Linnean species include several subspecies or 

 " elementary species " or " varieties," just as there are races 

 and breeds of domesticated animals and cultivated plants. In 

 studying the peculiarities of a whitlow-grass {Draba verna), 

 or of a wild heartsease [Viola tricolor), or of a pigeon, or of 

 a rabbit, we must, obviously, estimate these peculiarities in 

 reference to the particular stock or race to which they belong. 



On the other hand, many individuals differ markedly from 

 the mean of the stock or race to which they belong. In some 

 character or characters they are extraordinary individuals. 

 What is the chief conclusion in regard to the offspring of these 

 individuals ? It is that they are, on an average, more mediocre 

 than their parents. 



As Mr. Yule puts it, " This phenomenon of the relapse of the 

 offspring from the parental type towards mediocrity is termed 

 regression. Regression and not constancy of type is, for the 

 statistician, the fundamental phenomenon of heredity and the 



21 



