MODIFICATION OF THE GERM-CELLS IN MAMMALS 151 



the members of the large-size Utters than among those in the 

 small litters, and the large litters are 1.7 times more frequent 

 in the control than in the alcoholic lines. 



The mortality is corrected on the basis of the normal records 

 as follows : The rate for the normal animals born one in a litter is 

 zero; two in a litter, 15.21 per cent; three in a litter, 16.66 per 

 cent; four in litter, 37.5 per cent, and five in litter, 20 per cent. 

 On this basis what should be the number of alcoholic animals 

 dying in the several different-size litters? The numbers should 

 be zero instead of 7 for one in litter animals; 24.64 instead of 30 

 for two in litter animals; 46.48 instead of 109 for three in litter 

 animals; 37.5 instead of 52 for individuals born four in litter, and 

 3 instead of 13 for five in litter. These numbers give a total of 

 111.62, which divided by the number of alcoholic animals, 594, 

 shows a mortality percentage of 18.79. On the basis of the 

 control mortality for the different-size litters, this is what the 

 mortality should have been in the alcoholic lines, yet instead of 

 18.79 per cent it was actually 35.52 per cent, or almost double 

 the normal rate. Again to express the corrected mortality in 

 the alcoholic lines in terms of the control as 100, we find that for 

 every 100 of the control animals that die 189 from the alcoholic 

 lines die. 



The last column shows the 302 alcoholic inbred animals to 

 present a still worse record. The actual mortality here is 39.07 

 per cent, or one and three-fourths times higher than in the con- 

 trol. Here again correcting as in the preceding cases, the mor- 

 tality on the basis of the control record in the different-size 

 litters, it should normally be 18.59 per cent, but instead the 

 mortality is 2.1 times greater than this among these alcoholic 

 inbred animals. In other words, for every 100 control animals 

 that die 210 alcoholic inbred individuals succumb. While the 

 normal inbred animals, although their numbers are small, pre- 

 sent a slightly better record than the straight control, 98 of 

 these dying to 100 of the control. 



In the third and fourth horizontal spaces of the table the total 

 mortality is divided into the prenatal and postnatal deaths. 

 The proportion of prenatal to postnatal death in the different lines 

 presents peculiar arrangements that will be seen to exist, not only 



