794 



AMERICAN MEX OF SCIENCE 



compared with that of their parents, and 

 confirms the common belief in the tend- 

 ency to an extinction of the families of 

 men who work hard with the brain. ' ' Mr. 

 Havelock Ellis^ found that 214 fertile 

 marriages of British men of distinction 

 produced, on the average, 5.45 children, 

 while 276 "genius producing families" 

 consisted, on the average, of 6.96 children, 

 and remarks that "men of genius belong 

 to families in which there is a high birth 

 rate, a flaring up of procreative activity." 

 He says further that this "might, indeed, 

 have been anticipated. The mentally ab- 

 normal classes generally belong to families 

 with a high birth rate ' ' ; and quotes data 

 in regard to criminals and the insane. 

 Thus two wide generalizations that in- 

 tellectual performance conduces to sterility 

 and that genius is allied to insanity are 

 based on a curious statistical fallacy for 

 which one would suppose Galton the least 

 likely of men to be responsible. 



In a population whose families remain 

 of the same average size in successive gen- 

 erations, every one, whether he be a scien- 

 tific man, a criminal or a tailor, is likely 

 to come from a larger family than he has. 

 If, for example, all families were of one or 

 of seven children equally divided, the aver- 

 age family would be four in each genera- 

 tion, but the children would be seven timas 

 as likely to come from the larger family 

 and would belong to a family which, on 

 the average, would be 6.25. With an aver- 

 age family of three, the size of 100 fam- 

 ilies would be distributed approximately 

 as follows: 



Size ..0 123456789 10 

 No. .. 18 10 20 23 15 7 5 3 2 1 1 



When then we count up the average 

 size of the family from which each of the 

 300 children comes, it will be found to be 

 4.47. As our scientific men come from 

 families of tlie average size of 4.7, one may 

 conclude that the families of the class to 

 which they belonged were of the average 



8"A Study of British Genius," London, 1904. 



size of about 3.2. In one generation this 

 family has been reduced to 2.3, owing 

 either to a general fall in the birth rate 

 or because scientific men have families 

 which are smaller than those of the classes 

 to which their parents belonged. Both 

 factors are present ; there is a general 

 decrease in the birth rate and the edu- 

 cated classes have families smaller than 

 the average. 



In Table VI. is given information in 

 regard to the size of family of the parents 

 of scientific men in accordance with their 

 nationalities and their occupations. The 

 Germans had families of the average size 

 of 5.7, the British of 4.8 and the native 

 Americans of 4.5. The probable errors 

 show that these differences are not due 

 to the limited number of cases. It is 

 known that immigrants from foreign na- 

 tions have larger families than native 



Table VI. The Size.s of the Families of the 

 Parents of the Scientific Men in ac- 

 cordance WITH THEIR Nationalities 

 and their Occupations 



Total , 



871 



4.65 



Professions 



Clergj-men 



Physicians 



Lawyers 



Teachers 



Others 



Agriculture 



Manufacturing and trade. 



381 



88 

 65 

 60 

 75 

 93 



185 

 311 



4.51 



4.77 

 4.22 

 4.58 

 4.39 

 4.54 



5.09 

 4.59 



Total , 



877 



4.66 



Americans, but these figures probably give 

 the only information in regard to the fam- 

 ilies which produce scientific and other 

 professional men. Tlie disparity is not so 

 great as in the so-called lower classes, but 

 it is sufficient to indicate that in the pro- 

 fessional classes descendants of recent im- 

 migrants from Germanv and Great Britain 



