58 Life and Letters of Francis Galton 



Natural Inheritance may be antiquated now, but in the history of science 

 it will be ever memorable as marking a new epoch, and planting the seed 

 from which sprang a new calculus, as powerful as any branch of the old 

 analysis, and valuable in just as many fields of scientific research. 



In its application to inheritance the work suffers from the same misinter- 

 pretation of "regression" that I have several times referred to, namely making 

 the regression of offspring of given parentage a great biological law, when 

 it really arises from the clubbing together of all offspring of given parentage 

 without regard to their earlier ancestry. Given selected parentage and grand- 

 parentage alone, then with our present numerical values of the multiple 

 correlation constants, it seems highly probable that the progeny of selected 

 offspring would progress rather than regress on their parents and grand- 

 parents. In other words, given a line in which by chance or artificial selection 

 there has been marked ancestry for two or three generations, and which is 

 then isolated or inbred, there is reason to believe it would progress even 

 beyond its ancestry rather than regress. Statistical investigations of heredity 

 since 1889 seem to indicate a progressive evolution in selected lines, rather 

 than a general regression to a population mean*. That would only arise from 

 the far too frequent mating with mediocrity or worse than mediocrity. If 

 Galton's misinterpretation of regression runs through Natural Inheritance, 

 and makes him appeal to "sports" for evolutionary changes; if the reader 

 is puzzled to know why Galton should study "variations proper," which 

 according to him have no permanent value for evolution ; still the book is a 

 great book, for it applies a wholly new calculus — if one still in its infancy — 

 to an important biological problem. 



I think, however, that Galton fully grasped how much more important 

 was his method than its special application. He writes that his conclusions 



"depend on ideas that must first be well comprehended, which are now novel to the large 

 majority of readers and unfamiliar to all. But those who care to brace themselves to a sustained 

 effort, need not feel much regret that the road to be travelled over is indirect, and does not 

 admit of being mapped beforehand in a way that they can clearly understand. It is full of 

 interest of its own. It familiarises us with the measurement of variability, and with curious laws 

 of chance that apply to a vast diversity of social subjects. This part of the inquiry may be said 

 to run along a road on a high level, that affords wide views in unexpected directions, and from 

 which easy descents may be made to totally different goals to those we have now to reach. I 

 have a great subject to write upon, but feel keenly my literary incapacity to make it easily in- 

 telligible without sacrificing accuracy and thoroughness." (Chapter I, pp. 2-3.) 



Galton in his Introductory Chapter states that there are three problems 

 with which he will be principally concerned. The first problem is to deter- 

 mine how a population can, under the laws of heredity, keep stable from 

 generation to generation. The second problem regards the average share 

 contributed to the character in the offspring by each ancestor severally. 

 The third problem is to measure numerically the nearness of kinship in 



* There has always been this element of truth in Johansen's theory of "pure lines," that 

 selected lines do not regress if they are isolated or inbred. The doubtful dogma of that theory 

 is that exceptional members of a "pure line" are only "fluctuating variations," and so no further 

 selection is of any value within a "pure line." 



