190 



A. FRANKLIN SHULL. 



about the same age, was isolated to determine the sex ratio of 

 the second generation, and so on. 



From 30 to 100 daughters, all of nearly the same age, and from 

 the same parts of their respective families (end of first day of 

 egg-laying) were recorded for each generation. A new genera- 

 tion was, by this method, secured every three days, as a rule, 

 instead of every two days, as is possible when the first daughters 

 of a family are used as parents of the next generation. 



The number of male-producing and female-producing females 

 in each generation thus recorded is given in Table I. Male pro- 



Table I. 



Showing the Nxjmber of Male- Producing (d^ 9) and Female- producing 



( 9 ? ) Females in a Line of Hydatina senta. 



The male-producers occur periodically. 



duction in this line, when bred as described above, showed a 

 well-marked periodicity, the interval being about one month. 

 Conditions were purposely kept fairly uniform; but even if it 

 were not possible to prevent changes of the medium, it is scarcely 

 probable that any external condition favoring male production 

 should have recurred with such regularity. Moreover, were this 

 periodicity due to external factors, the intervals between periods 

 of male production should be the same in all lines bred at the 

 same time and subjected to the same conditions. That this 

 was not the case will now be shown. 



