370 



HELEN DEAN KING 



age of about twenty-two litters to the generation : in the last nine 

 generations the average was only about three litters to a genera- 

 tion. Such a small number of litters as that produced in the 

 later generations of this series does not afford an opportunity for 

 a careful selection of breeding stock, neither does it furnish 

 sufficient data to make the results of statistical value. 



In the successive generations of mice bred by von Guaita there 

 was, to quote Davenport ('00) : "a, reduction in fertility of about 

 30 per cent, and this is probably due to close inbreeding." In 

 order to make this deduction from von Guaita's data, however, it 



TABLE 14 



Showing the number and average size of the litters in twenty-nine generations of white 

 mice inbred by August Weismann, and in seven generations of hybrid mice 

 inbred by von Guaita 



is necessary to combine the records for three generations, as 

 Davenport did. If the records for the various generations are 

 considered separately, or grouped by twos, there is not the steady 

 decrease in fertility with advancing inbreeding that Davenport's 

 grouping of the data implies. Taking the data for the first two 

 generations together, the average size of the litters was 3.7; for the 

 next two generations there was an average of 4.0 young per litter ; 

 in the final group the litters averaged 2.7 young. Since the 

 crossing of varieties is supposed to increase vigor and fecundity 

 it seems strange that the Fi and the F2 litters in this series should 

 contain a smaller average number of young than is found in the 



