THE MENDELIAN CLUE 231 
the inheritance of particular characters through 
successive generations and to measure quantitatively 
the degrees of hereditary resemblance. He was 
led to the law of ancestral inheritance, according 
to which the average contributions to each inherited 
faculty are a half from the parents, a quarter from 
the grandparents, an eighth from the great-grand- 
parents, and so on backwards in the same diminish- 
ing ratio (afterwards somewhat modified by Pear- 
son); and another deduction was the law of filial 
regression or the tendency to approximate to the 
mean of the stock. It is necessary, however, to bear 
in mind that these average statistical deductions do 
not hold in regard to non-blending hereditary 
characters, and that they do not seem to take 
sufficient account of the fundamental distinction 
between inborn variations and individually acquired 
modifications. The latter are somatic dints due to 
peculiarities of nurture, and have not been proved 
to be transmissible; the former are expressions of 
germinal changefulness, and are in some cases 
demonstrably transmissible. Of great value, how- 
ever, has been the statistical demonstration of the 
heritability of subtle constitutional qualities—such 
as fecundity and longevity—and the proof that 
clearly-defined mental qualities may be handed on 
to, and distributed among, the offspring just in 
the same way as bodily characters. 
The first year of this century will be memorable 
in the annals of biology for the rediscovery of the 
Mendelian clue by Correns, De Vries, and Tscher- 
