Nos. 622-623] THE SEXES IN THE APHID 



519 



tioned, and notwithstanding the slower development, 

 smaller young and smaller daily output of young of the 

 alate females, these winged individuals be assumed to 

 have a higher rate of metabolism than the wingless ones, 

 other difficulties are encountered. This assumption 

 would have the advantage of allowing parent and off- 

 spring to be of opposite metabolic type in both the par- 

 thenogenetic and sexual portions of the cycle, instead of 

 being of opposite type in the parthenogenetic phase and 

 of like tvpe in the sexual phase. The transitions, how- 

 ever, would be in opposite directions in different parts of 

 the cycle. In the parthenogenetic portion there would 

 be a transition from a low rate of metabolism to a high 

 rate (wingless to winged) ; while in the sexual part of 

 the cycle the transition would be from high rate to low 

 rate (male to sexual female). These opposite transitions 

 would have to occur in part simultaneously, as m Table 

 IV, fourth and fifth generations. 



Unfortunately there has been no opportunity to deter- 

 mine experimentally the rate of metabolism in the various 

 kinds of individuals in Macrosiphum; that is part of the 

 program for the future. In the meantime, whether there 

 is a -fallacy in the foregoing argument, or a fallacy in 

 Riddle's conception of the relation of metabolism to sex, 

 can not be asserted with any degree of confidence. 



Obviously the mere rate is not the only feature of metab- 

 olism that may conceivably be related to sex. If there 

 are qualitative differences in the reactions that consti- 

 tute metabolism, it seems to me more likely that these 

 would influence the development of sexual organs than 

 that the production of ovaries rather than testes could 

 be determined by rate of metabolism alone. Qualitative 

 differences in the reactions might entail differences m the 

 rate of GO, production, and therefore be interpreted as 

 quantitative differences. An increase in the output of 

 lumber from a sawmill might be taken to indicate that 

 the saws were running faster than formerly, whereas m 

 reality the saws had been replaced by a new type of saw. 



