EAR-SIZE 



27 



Mating 2 shows a result more nearly normal. Two individuals exceed 

 the mid-parental ear-length by 3 mm., 2 fall short of it 7 mm., and i by 

 9 mm. The total range of variation is 12 mm. 



In mating 3, which jjroduced 2 litters of young, the variations are again 

 chiefly below the mid-parental ear-length, but to no greater extent than 

 we might expect, in view of the difference in age between parents and 

 children, when measured. In litter i the deviations are —2, +3, —4, 

 and —2 mm., a nearly normal result; but in litter 2, four individuals 

 show a deviation of —7 mm., and one a deviation of +3 mm. The range 

 in htter i is 7 mm.; in Utter 2, 10 mm. There is no evidence of hetero- 

 geneity among the gametes. 



Table 15. 



• At 20 weeks. 



Some other matings which fall in the category of cross 6, and consti- 

 tute a second generation (Fj) from cross 3, are included in table 16. 



The statistics contained in table 16 are not very satisfactory because 

 the observations are made at such dilTerent ages, and because, in one case 

 at least, that of (?38i, a remarkable increase in ear-length is recorded 

 subsequent to the age of 18 weeks. For observations made at the same 

 age, however, the variability in ear-length is considerable. Tlie range 

 of variation in mating i, litter i, is 17 mm.; in litter 2 it is 15 mm.; in mating 

 2, it is 18 mm. In generation F,, cross 3, the range of variation was only 



