44 HELEN DEAN KING 



less than fraternal variability in stock animals. The evidence as 

 given indicates that the inbred males were more variable in body 

 weight than the males in a single stock litter up to 212 days of 

 age, but that at all subsequent ages stock males were very 

 much more variable. Omitting from the inbred series the co- 

 efficients for the first two age periods, since there were no corre- 

 sponding coefficients for the males of the stock litter, the average 

 coefficient for the males in the fifteenth inbred generation, tak- 

 ing all ages together, was 7.8, while the average coefficient for the 

 stock males was 9.1. The difference between the coefficients was 

 not very great, but it seems large enough to signify that the 

 variability in the body weights of the males in the fifteenth in- 

 bred generation was somewhat less than fraternal variability in 

 the stock controls. 



The relative variability of inbred and stock females was 

 slightly different from that found in the male groups. Females 

 of the fifteenth inbred generation were, as a rule, more variable 

 in body weight than females in a single stock litter up to 120 

 days of age, beyond this age stock females seemed to be the 

 more variable. The difference of 0.2 points between the aver- 

 age coefficients for the two groups, omitting the findings for the 

 first two weighings of the inbred group, was in favor of the stock 

 litter. This difference, however, is much too small to have 

 any meaning, and it is evident that the variability in the body 

 weights of the females of the fifteenth inbred generation was 

 about the same as fraternal variability in the stock controls. 



4. DISCUSSION OF RESULTS 



This study of the growth in body weight of albino rats belong- 

 ing to fifteen generations produced by brother and sister mat- 

 ings has shown that the closest form of inbreeding possible in 

 mammals does not necessarily produce animals that are below the 

 normal body size, as Crampe ('83) and Ritzerfla-Bos ('93; '94) 

 have maintained. During the early part of these experiments 

 all of the evil effects that are said to follow from close inbreed- 

 ing were obtained, but it was shown conclusively that they 



