A STUDY OF BRITISH GENIUS. 599 



be abnormally large.* In genius-producing families there is an in- 

 variable deficiency of families below the average normal size, and an 

 invariable excess of families above that size. In the largest size group 

 (over 14) the excess becomes extravagantly large; this, however, may 

 be partly accounted for; we may be sure that the biographers have 

 seldom failed to record families of this size, so this group has really 

 been recruited from the families of all our 902 eminent persons. Even 

 on this basis, however, it remains extremely large; in Denmark, for 

 instance, it is stated, a family of 22 children only occurs once in 34,000 

 marriages.f 



If, as seems probable, it may be asserted that genius-producing fami- 

 lies are characterized by a tendency to an abnormally high birth-rate, 

 this is not a fact to cause surprise. It might, indeed, have been antici- 

 pated. The mentally abnormal classes generally belong to families with 

 a high birth-rate. This has been shown by Ball and Kegis (confirmed 

 by Marandon de Montyel) to be markedly the case as regards the insane. 

 Magri has found it to be the case as regards criminals, as well as regards 

 the epileptic, hysterical and neurasthenic. 



An interesting point, and one which can scarcely be affected at all 

 by any twist in the biographical mind, is the fact that our men of 

 ability (the women are here excluded) are the offspring of predominantly 

 boy-producing parents. Taking the 64 families in which the number 

 of boys and girls in the family is clearly stated, and excluding 12 of 

 these as consisting only of boys, we find that there are about 

 6 boys to 5 girls, or more exactly, 111 boys to 100 girls. The 

 normal proportion of the sexes at birth at the present time in England 

 is about 104 boys to 100 girls. It is in accordance with the pre- 

 dominantly boy-producing tendency of families yielding men of genius 

 that the families yielding women of genius should show a predominantly 

 girl-producing tendency. Here, indeed, our cases are far too few to 

 prove much, but the results are definite enough so far as they go. Put- 

 ting aside the families consisting only of girls, the sexual ratio is rather 

 more than 3 boys to 4 girls, or more exactly, in the ratio of 85 boys to 

 100 girls. Putting the matter in another way, we may say that, while 

 in every ten families from which men of genius spring, the boys pre- 



*This tendency has already been noted by Galton when investigating English 

 men of science, and by Yoder in studying a small miscellaneous group of eminent 

 men. 



fin our genius-producing group there are 4 families of more than 19 children. 

 Doddridge was the youngest of 20 children; Popham was the youngest of his 

 mother's 21 children; Colet was the eldest and only surviving child of 22; Demp- 

 ster was, or stated himself to be, the 24th, of 29 children. There was a strong 

 tinge of romance about Dempster, and 1 fear that we cannot accept this statement 

 with such complete confidence as would be desirable. 



