520 



THE AMERICAN NATURALIST [Vol. LII 



So long as rate of metabolism can be determined experi- 

 mentally, while the precise reactions can not, there is 

 every reason to continue the attempt to relate the rate 

 of reaction to the course of development. But when facts 

 come to light which do not easily fit preconceived ideas, 

 it is highly important that alternate possibilities be kept 

 in mind. 



It is not impossible that the difficulties discussed above 

 may be removed by discovering that the metabolic change 

 that causes the transition from wingless to winged females 

 is different from the change that causes the transition 

 from males to sexual females. The two changes may be 

 more or less independent of each other. In that case it 

 may be possible to separate them experimentally. An 

 agent may sometimes be found which will hasten or post- 

 pone the sexual reproduction without in any way affect- 

 ing the transition from wingless to winged females in the 

 parthenogenetic phase. If this agent hastened the sexual 

 reproduction, it should act as a male-producing factor, 

 since sexual forms would be introduced while wingless 

 parthenogenetic females were more abundant. If, on the 

 other hand, the agent delayed sexual reproduction, it 

 should favor females, since the parthenogenetic mothers 

 would then be more largely winged. 



BIBLIOGRAPHY 



Gregory, Louise H. 



1917. The Effect of Starvation on the Wing Development of Micro- 



siphum destructor. Biol. Bull, Vol. 33, No. 4, October, pp. 



296-303. 

 Riddle, Oscar. 



1917. The Theory of Sex as Stated in Terms of Results of Studies on 



