156 CHAELES R. STOCKARD AND GEORGE N. PAPANICOLAOU 



ample, an animal derived from a directly alcoholized father 

 and a normal mother could be said to contain half affected and 

 half normal germ plasm, whereas another in whose pedigree the 

 only alcoholic individual was an alcoholized grandfather, would 

 undoubtedly contain a smaller amount of affected stuff. 



Finally, then, in the light of the facts involved, the general 

 table presents an impression closely similar to that derived 

 from the previous records of these experiments, but it adds data 

 of much importance for a clearer understanding of the problems 

 concerned. 



The improvement in the present records over the former ones 

 might suggest that should the methods of breeding and caring 

 for the animals reach perfection, the differences between the 

 alcoholic lines and the control might be entirely erased. This 

 would be possible if the improvement was due alone to method, 

 but such a suggestion ignores the fact that the improvement is 

 more largely due to the presence of late-generation animals with 

 only a small amount of alcoholic germ plasm in their ancestry 

 and a large number of normal progenitors. The analysis of the 

 following table 2, in which the several generations are treated 

 separately, will fully substantiate the validity of the foregoing 

 statement. 



Before considering this table, however, we may discuss briefly 

 the phenomenon of absorption of embryos in utero and our 

 methods of examining pregnant females in order to fully record 

 the fate of all embryos that begin to develop. A knowledge of 

 this prenatal mortality is involved not only in the table just 

 studied, but in several of those that follow. 



6. ABSORPTION OF EMBRYOS IN UTERO AND ABORTIONS OF PARTS 

 OF LETTERS : METHODS OF DETECTING THESE PROCESSES 



After having observed the course of pregnancy and the size of 

 the litters produced in a large number of cases, we became con- 

 vinced that many of the small litters delivered at full term were 

 only partial litters. Particularly in the alcoholic lines it became 

 evident that abortions of one or two members of a litter might 



