MODIFICATION OF THE GERM-CELLS IN MAMMALS 153 



later embryonic or uterine life. Furthermore, many individuals 

 die very soon after birth, while those that happen to survive the 

 periods shortly following birth are often capable of an almost or 

 quite normal existence. 



The mortality in the control is low, but half of this, or a high 

 proportion, occurs after birth. The mortality in the alcoholic 

 lines is high, but only a low proportion, about one-third of this, 

 occurs after birth. It may be added further that the young 

 alcoholics which die after birth in the majority of cases die 

 within a few days, while the control young that die after birth 

 are more likely to be scattered along over a number of days 

 or weeks. 



It is thus seen that in both the alcoholic and alcoholic inbred 

 lines there is a decided tendency for the developing embryos and 

 young to succumb during the early periods of their development. 

 This would suggest that these affected individuals were often 

 incapable of passing through the early critical stages of uterine 

 life. But if they were sufficiently fit to survive these periods, 

 their chance for existence was good, so that their postnatal 

 mortality, although actually higher than the control, was pro- 

 portionally much lower. Thus we have a somewhat rigid in- 

 dividual selection taking place during the stages of uterine life, 

 so that the sum total of the individuals at a given stage is of 

 a better average quality than during any previous stage and 

 vice versa. Therefore, as is clearly shown beyond, those ani- 

 mals of the alcoholic lines which live to become mature and 

 prove to be fertile are a strictly selected few and in each gen- 

 eration the proportion of strong to weak individuals through 

 this selection constantly tends to increase. 



The sixth horizontal space shows a complete absence of de- 

 fective individuals in either the normal or normal inbred groups. 

 It may be stated here that during the entire seven years of this 

 experiment not one grossly defective or deformed individual has 

 appeared in the nonalcoholic or control lines. This is a rather 

 remarkable record for any group of animals, and it speaks strongly 

 for the perfection of the original stocks from which both the 

 control and the alcoholic lines have been derived. 



