26 Eugenics Record Office, Bui.i.etin No. 4 



do is to compare the proportion of epileptics in a collection of 

 families in the latest generation (selected in nearly all cases be- 

 cause they have at least one epileptic in the fraternity) with their 

 paternal and maternal fraternities (selected insofar as parents, 

 uncles and aunts of an epileptic fraternity usually comprise at least 

 one epileptic). The required result is easily got from the tables. 

 It appears that of 1,020 persons belonging to the parental genera- 

 tion 124 or 12.2 per cent, are epileptic or imbecile; while of the 

 739 persons in the following or filial generation 270 or 36.5 per 

 cent, are epileptic or imbecile. It might be concluded, in other 

 words, that, given an ancestry such as is producing defectives, it 

 tends to produce offspring with three times the proportion of 

 defectives that it has itself. Such a conclusion would be unwar- 

 ranted. For in the same generation with the affected fraternity 

 are other fraternities of cousins none of whom contain defective 

 children; it is certain that if these unaffected fraternities were 

 included they would much reduce the proportion of defectives in 

 the younger generation. We have calculated the proportion of 

 epileptic and imbecile children to all children in the later genera- 

 tion and compared it with the ratio of the generation to which the 

 parents belong. In the earlier generation the relation is 159 defec- 

 tives out of 1,354 children; in the later generation, 366 out of 

 1,460. Or the ratios are, respectively, 11.7 per cent, and 25.1 

 per cent. That is, the proportion of defectives in the younger 

 generation is more than double that of the earlier. Making a 

 slight allowance for the possibility that the defects of some of 

 the parental generation have been concealed (and this, on account 

 of the inaccessibility of the parental generation the field worker 

 can not always check) it is probable that the proportion of defec- 

 tives in the population we are studying doubles in each generation 

 and is now i in each fraternity of 4! Were the population to 

 mate in the future about as in the past it is easy to see that, in 

 half a century, on the average over half of each fraternity of the 

 epileptic-producing population would be defective. 



Of course, the families constitute a small, selected portion of 

 the population of New Jersey. Probably the total number of 

 persons now living in New Jersey whose mental condition is ade- 

 quately described in our data is not over 2,500 or i to 1,000 of 

 tJie total population of the state. Our strains, then, constitute an 

 unimportant part of the population. Is there any reason to sup- 



