Xo. 636] GUINEA PIGS AND TUBERCULOSIS 39 



Returning to Tables VII, VIII and IX, there are no 

 differences between experiments CO (parents unrelated 

 inbreds), CA (sire crossbred, dam inbred) and AC (sire 

 inbred, dam crossbred), which indicate an influence of 

 the breeding of the dam on the resistance of the young. 

 There are rather large differences, it is true, but these 

 are not consistent and must be attributed largely to 

 heredity from particular families rather than to the 

 system of breeding. 



Differences in Resistance among Inbeed Families 

 Passing now to a comparison of the different inbred 

 families with each other, we come to results which ap- 

 pear much more striking than the differences between 

 inbreds and crossbreds. In all three groups of experi- 

 ments, one family, 35, stands out as distinctly more re- 

 sistant than the others. It leads the average of the 

 others by 16 per cent, in lots 3-5, 19 per cent, in lots 7-9 

 and 43 per cent, in lots 10-14. The more striking result 

 in the last case is probably due to the weaker inoculation. 

 In spite of the large amount of variation in each case, 

 the probable errors put the superiority of Family 35 

 beyond question. Among the other families, family 2 

 is on the whole the most resistant. Families 13 and 32 

 are about equally susceptible. Only one animal from 

 family 39 was tested. This one was one of the first in 

 its lot, no. 7, to die, indicating low resistance in this 

 family also so far as conclusions can be based on such 

 a slender basis. 



Inheritance of Resistance among Crossbreds 

 If there are hereditary differences in resistance, one 

 might expect to find differences among the crossbreds de- 

 pending on the families which went into their ancestry. 

 A preliminary test of this point was made as follows: 

 The length of life of each crossbred {CO, CA or AC) 

 was entered under each of the four grandparental inbred 

 families. Thus an animal in experiment CA, whose sire 



