INHERITANCE IN HYDATINA 



65 



244 of these eggs, or 68.3 per cent, had hatched (table 14). The 

 variabiHty in hatching time is not great, hence the standard 

 deviation is low (a = 2.19). 



The reciprocal crosses described in the last two paragraphs 

 are approximately equal both in percentage of viability, and in 

 variability of hatching time, though their parents C X D and 

 D X C, were widely different in both characters. This supports 

 similar evidence from the inbreeding oi C X D and the inbreed- 

 ing oi D X C. The reciprocal crosses C X D and D X C were, 



TABLE 14 

 Shoioing number of eggs that hatched, out of 357 in [D X C) X (C X D), figure 1 



Percentage of viable eggs. 



68.3 



Standard deviation: a^ = 2.19 



after all, nearly equal. The inequality existed only in the egg 

 stage. When these eggs grew up, so to speak, to produce par- 

 thenogenetic lines, these lines were essentially equal in the via- 

 bility and variability of the fertilized eggs which they produced. 



Effect of selection for low viability in a hybrid line 



Li. Females of the parthenogenetic line derived from a low 

 viability parent in {C X D) X (C X D) were paired with males 

 of the same line January 16 to 18, 1912. From 40 such matings, 

 515 eggs were obtained. These were kept under observation 

 until February 22, at which time 149 eggs, or 28.9 per cent, had 

 hatched (table 15). Standard deviation of hatching time has 

 not been computed for any of the selection experiments, as it 

 was obvious. that no marked difference would be found. 



THE JOURNAL OF EXPERIMEMTAL ZOOLOGY, VOL. 15, NO. 1 



