142 Heredity. 



It must be conceded to Buckle that the question of psycho- 

 logical heredity is by no means one that can be treated with strict 

 scientific rigour; and there are many reasons why this is so. 

 Oftentimes in the course of this present work we have felt the 

 insufficiency of the argument, 'A distinguished father, a distin- 

 guished son therefore talent is hereditary,' whereas we ought to 

 be able to show that to a given mode of mental activity in the 

 progenitor corresponds precisely the same mode in the descendant, 

 or, at least, to say why this is not so. But this is too much to 

 require in the present state of psychology. 



This granted, if we revert to the essential point of Buckle's 

 objection, we find that in his view the cases of heredity are simply 

 fortuitous successions, such as are to be found whenever we com- 

 pare a great mass of facts. If we take from the registers of a 

 lottery the winning numbers through a long series of years, we 

 should probably find that there were occasionally the same succes- 

 sion of numbers, the result of mere chance. In this way, or 

 nearly so, Buckle explains cases of heredity. He reduces the 

 question to a calculation of probabilities. But this singular 

 hypothesis had already been answered by a mathematician. 



Maupertuis, after citing a case of sexdigitism which persisted 

 through four generations, adds : ' I presume no one would regard 

 sexdigitism as the effect of mere chance. But suppose we so 

 regard it, let us then see what is the probability that this accidental 

 variation in a parent will not be repeated in the descendants. In 

 the course of an inquiry made by me in a city of 100,000 inhabit- 

 ants, I found two persons marked by this singular anomaly. 



' Suppose a thing not very easy that three other cases escaped 

 my observation, and that we have a man with six fingers for each 

 20,000 souls ; the probability that his son or daughter will not be 

 born with six fingers is as 20,000 to i, and the probability that his 

 grandson will not have six fingers will be as 20,000 times 20,000 

 (or 4 millions) to i. Finally, the probability that sexdigitism will 

 not continue through three successive generations will be as 8000 

 millions to i, figures so large that the certainty of things best 

 demonstrated in physics does not approximate to these proba- 

 bilities.' 1 



1 Maupertuis, (Ettvra, voL ii. letter 17. 



