352 Aaron Franklin Shiill 



shown that the death losses are probably not selective; and even 

 if selective they are, in certain experiments, entirel)/ insignifi- 

 cant. Nothing has been said on this point regarding the starv- 

 ation experiments. If it could be shown that the shortening 

 of the families in these experiments were due to death of many 

 of the female-producers, a considerable increase in the propor- 

 tion of male-producers would be accounted for, and the experi- 

 ments might only show that when the rotifers were starved many 

 female-producers died. That the starved families were smaller 

 was due chiefly to the fact that fewer eggs were laid, and only in 

 small part to failure of the eggs to hatch or to death of the young 

 rotifers. But no amount of elimination of female-producers 

 could result in an increase m the absolute number of male-pro- 

 ducers. Such an increase in the number (as well as proportion) of 

 male-producers is found in the second and sixth parts of Table 

 IV, an increase too great to be insignificant. I conclude, there- 

 fore, that death losses do not vitiate the results of the starvation 

 experiments. 



Previous workers with Hydatina have spoken of the problem 

 which its varying proportion of male-producers presents, as one 

 of sex-determination; but it is open, as I have indicated in the 

 introduction, to interpretation as a change from the partheno- 

 genetic to the sexual phase of the life cycle. This view was adopted 

 by Morgan ('07, p. 346). The assumption necessary to support 

 this view was that the male-producers are the sexual females, 

 which assumption was based on the numerical relations found 

 by Maupas ('90b), the cytological evidence of Whitney ('09), 

 and the analogy afforded by Asplanchna (Lauterborn, '98, 

 p. 178). Since among many thousands of females laying only 

 eggs that develop parthenogenetically, not one has ever been 

 found to produce offspring of both sexes, my observation that 

 a male-producer may also lay resting eggs, though not a complete 

 demonstration, leaves little doubt that male eggs and sexual 

 eggs are identical. My observation does not exclude the possi- 

 bility that female eggs may also be fertihzed, but Maupas's 

 experiment mentioned in the introduction and the chromosome 

 counts made by Whitney make this improbable. 



