58 



A. FRANKLIN SHULL 



These inbred eggs are so different from those of C X C, both 

 in their percentage of viabihty and in the variabihty of the 

 duration of the egg stage, that reciprocal crosses between the 

 lines producing them (C and D) are of considerable interest. 

 These crosses are next described. 



C X D. Females of line C were mated with males of line D 

 October 14 to 16 and 20 to 21, 1911. From 55 such matings 

 were obtained 827 eggs. These were observed daily until Decem- 

 ber 4, at which time 426, or 51.5 per cent had hatched. More 

 than half the eggs were observed eight days longer, but none of 

 them hatched, so the 426 recorded up to December 4 are prob- 

 ably about the total. Table 5 gives the details of the hatching. 

 The eggs were obtained in two distinct lots, separated by an 

 interval of several days, and these lots were observed separately. 

 In the table, they are combined, and instead of the actual date 

 the day of hatching, counting from the day the first egg was 

 laid, is given in the table. From this table it appears that the 

 variability of the duration of the egg stage is larger than in 

 C X C, but much smaller than in D X D {a = 3.44.). 



A parthenogenetic line, started from a female selected at ran- 

 dom from the third day's hatch is recorded in table 6. The 

 proportion of male-producers in this line was 44.6 per cent. 



TABLE 5 

 Showing number of eggs that hatched, out of 827 in lot C X D, figure 1 



