EFFECTS OF INBREEDING AND CROSSBREEDING. 7 



tions, and produced its last litter in January, 1913, having reached 

 the ninth generation. The family characters may in the main be 

 considered to represent the large stock descended from Male 13. 

 In Family 14 the two lines kept about the same ratio to each other 

 in strength. The line from Male 2 was the strongest in numbers. 

 It ran out in the tenth generation, while the line descended from 

 Male 13 lasted only to the ninth generation. 



The families other than 13 and 14 all descend from a single original 

 pair but yet are of varying degrees of homogeneity. Some idea of 

 the degree of homogeneity of the various families can be obtained 

 from Table 2. This table shows the number of matings made in 

 each family in two periods, 1906 to 1909, inclusive, and 1910 to 1914, 

 inclusive, and the maximum percentage of these matings which can 

 be traced back to a single mating in each generation of inbreeding. 

 The original mating is called the zero generation. 



The table reveals that certain families, such as 18 and 39, became 

 dominated by a particular subfamily in their early history and 

 remained so later. At the other extreme are families, such as 11, 

 14, 17, and 32, which remained split up into many subfamilies even 

 through 1914. Most of the families became more homogeneous as 

 time went on. Families 2 and 38 are extreme examples of the 

 emergence of one subfamily into predominance. In a few cases 

 (19, 24, 31, and 35) the most important subfamily in the first period 

 became supplanted by another in the second. The most remarkable 

 case of this sort, that of Family 35, is not fully brought out by the 

 table. One of the four matings made in the second generation pro- 

 duced only 10 out of 59 matings of the third to seventh generations. 

 Its only pair of descendants in the seventh generation produced 49 

 out of the 75 matings of Family 35 of the eighth to twelfth genera- 

 tions and produced all matings following the twelfth. This family 

 has reached the twenty-third generation (1921) and now traces 

 entirely to a single mating of the twelfth generation. There seems 

 to have been no conscious effort to bring about the predominance 

 of the descendants of the single mating of the seventh generation. 



In interpreting the results in the various inbred families, it will 

 be important to bear in mind that those in which one subfamily 

 was predominant from the first or in which several subfamilies 

 remained about equally important should maintain about the same 

 average of hereditary characteristics throughout their history, while 

 a marked change in the hereditary characteristics of the family as a 

 whole need not be surprising in cases in which a subfamily, which is 

 unimportant at first, later emerges into predominance. 



