CASTLE. — INBREEDING, CROSS-BREEDING, AND SELECTION. 773 



ing the inbreeding experiment, the common ancestor of all the inbred 

 families, had twelve teeth in his right comb, ten in the left, numbers 

 higher than the normal and very similar to those prevailing among the 

 inbred descendants of the male in question. 



The variability in number of teeth has apparently been neither in- 

 creased nor diminished by the inbreeding. 1 The inbred flies have a 

 slightly higher standard deviation, as we should expect from the fact 



(Teefh 7 



Figure 5. Variation in number of teeth on the sex-comb of forty males from each 

 of three different broods (A, B, and C) of a stock inbred for six generations, 

 and of males (N) from a stock not inbred. 



that they have higher means, but the coefficient of variability (ratio of 

 standard deviation to mean) is in family B a little greater than in the 

 normal flies, while in families A and C it is a little less. 



The number of teeth is about the same, and varies in a similar way on 

 the two sides of the body, as is shown by plotting separately the numbers 



1 An examination was subsequently made by A. H. Clark of the variability of 

 the sexual comb in 45 males of the fifteenth inbred generation. This was found to 

 be almost identical in character with what it had been in the sixth generation. 

 The range of variation was still from 8 to 13, with the mode at 11, the mean 

 at 10.79. 



