Eugenics as a Creed and the Last Decade of Galton's Life 229 



"They found great industries, establish vast undertakings, increase the wealth of multitudes, 

 and amass large fortunes for themselves. Others, whether they be rich or poor, are the guides 

 and light of the nation, raising its tone, enlightening its difficulties, and imposing its ideals. 

 The great gain that England received through the immigration of the Huguenots* would be 

 insignificant to what she would derive from an annual addition of a few hundred children of 

 the classes W and X. I have tried but not yet succeeded to my satisfaction, to make an 

 approximate estimate of the worth of a child at birth according to the class he is destined to 

 occupy when adult. It is an eminently important subject for future investigators, for the 

 amount of care and cost that might profitably be expended in improving the race clearly 

 depends on its result." (p. 528.) 



Thus far it will be clear to the reader that all that Galton does is to assert 

 and assert with truth that in any scale of civic worth, whether it be one of 

 brains or energy, artistic power or skill, the classes W and X are of the 

 highest value to a nation, and should be multiplied if possible, the classes t, u, v 

 and below are undesirable, and should be decreased if feasible. It is difficult 

 to see how anyone can deny this, for by the very definitions of those classes 

 they are the best and the worst in the community. 



Galton now passes to " the descent of qualities in a population." Here he 

 makes use of the conception of regression as he has discussed it in his Natural 

 Inheritance, and makes the parental correlation one-third. As in that work 

 he indicates with a diagram how a population reproduces itself. The same 

 criticism may be made here as earlier on our pp. 18, 23 and 65, namely in the 

 first place the parental correlation is actually .much higher than he assumes 

 it, and secondly he supposes the ancestors of the parents in all cases to be 

 mediocre, whereas these ancestors will most probably deviate from mediocrity 

 in the same direction as the parents themselves do. Luckily these slips do 

 not invalidate his conclusions, for, if corrected, his case for obtaining F-class 

 offspring most economically by encouraging parentage in V-, U-, or T-class 

 individuals is greatly strengthened. If the reader will bear in mind that 

 Galton's statements owing to the above reasons give results far less favour- 

 able than they should be to F-class parents, we need not hesitate to cite his 

 sentences on p. 531 : 



"Of its [the T-class in new generations] 34 or 35 sons, 6 come from V parentage, 10 from U, 

 10 from T, 5 from S, 3 from R, and none from any class below R; but the number of the 

 contributing parentages has also to be taken into account. When this is done, we see that 

 the lower classes make their scores owing to their quantity not to their quality, for while 

 35 F-class parents suffice to produce 6 sons of the F-class, it takes 2500 .ff-class fathers to 

 produce 3 of them. Consequently, the richness in produce of F-class parentages is to that of 

 the ^-class in an inverse ratio, or as 143 to 1. Similarly the richness in produce of F-class 

 children from parentages of the classes U, T, S, respectively, is as 3, 11 '5 and 55 to 1. More- 

 over nearly one-half of the produce of F-class parentages are V or U taken together, and nearly 

 three-quarters of them are either V, U, or T. If, then, we desire to increase the output of 

 F-class offspring, by far the most profitable parents to work upon would be those of the F-class, 

 and in a three-fold less degree those of the fAclass." (p. 531.) 



Here we see Galton fully cognizant of the solution of the paradox which 

 nearly thirty years later was still troubling the non-statistical mind of 

 Professor Leonard Hillf . 



* This is an illustration often used by Galton, e.g. in his Presidential Address to the 

 Demographic Congress, 1891, and in the Jewish Chronicle, July 30, 1910. 

 t See this volume, p. 27. 



