BETTER AMERICAN FAMILIES III 



Showing How the Level of a Trait or Trait-Complex May Be Raised 

 Through Marriage-Selection 



WiLHELMINE E. KeY 



Eugenics Record Office, Carnegie Institution of Washington, Cold Spring Harbor, 



Long, Island, N. Y. 



IN tracing the history of the Rufer 

 family as given in the preceding 

 paper, the conclusion seems justified 

 that through selective matings, the 

 grade of such traits as aggressiveness, 

 perseverance and the ability to handle 

 number was noticeably increased or 

 diminished. For example, in Line E 

 the absence of such abilities with the 

 concomitant failure of the individuals 

 to marry those possessing the abilities in 

 question persisted through five genera- 

 tions. In Line C a comparatively low 

 ability to calculate has, through selec- 

 tive matings with good or average 

 ability, been built up into ability which 

 is average or in some instances even 

 above the average. This holds true 

 also of the trait-complex • aggressive- 

 ness. In Line D again, ability with 

 number, and aggressiveness have 

 through matings with a low grade of 

 these abilities, persisted at the low 

 level of the original founders. In the 

 same way the trait-complex persever- 

 ance is found to move up or down in 

 fair accord with the type of mating 

 made. 



Let us now see how far the behavior of 

 these abilities in inheritance conforms 

 to the Mendelian hypothesis. At the 

 outset, we concede that our data, since 

 they appertain to mental and tempera- 



mental traits, do not permit of the 

 accuracy of measurement possible in 

 handling physical traits. It will, how- 

 ever, be generally agreed that all have a 

 basis in heredity and remain far less 

 affected by environmental influences 

 than many other characters that might 

 be selected for study. And with at 

 least one ability, that of dealing with 

 number, it is possible to adhere to 

 unifonn objective standards. Our jus- 

 tification for attempting to carry the 

 Mendelian principles into this domain 

 is the value which even approximate 

 conclusions may have for future effort 

 along similar lines. 



THE ABILITY TO CALCULATE 



If we let C stand for comparative 

 presence of determiners for this abilit}^ 

 and c for their comparative absence, 

 whether they occur in a unit-like series 

 of similar or various elements, we have 

 the following possible combinations in 

 the union of genn plasms : CC develop- 

 ing relatively high ability which we may 

 indicate by H; Cc developing mediocre 

 ability indicated by M ; and cc develop- 

 ing low ability indicated by L. 



Distribution of these abilities from 

 different types of mating would be as 

 follows : 



cc X cc — 100% cc, i. e., All children show very low ability. 



Cc X cc — 50% Cc and 50% cc. Half the children show medium ability, half low 



ability. 

 Cc xCc— 25% CC, 50% Cc and 25%, cc. One-fourth the children with high, half 



with medium and one-fourth with low ability. 

 cc X CC — 100% Cc. All children show medium ability. 

 Cc X CC— 50%f, Cc and 50%o CC. Half show medium ability and half high. 

 CC X CC— 100% CC. All children show high ability. 



107 



