1 6 J. H. Powers 



before sexual development as such begins. I have not ascer- 

 tained this by measurements because of the difficulty of determin- 

 ing the sex of immature specimens. The difference, however, is 

 not great, and not only do exceptional individuals occur, very 

 slender larvae becoming females and robust individuals males, but 

 under certain special conditions of development the general law 

 is set aside altogether, excessively robust larvae producing, in the 

 rule, males and not females. This general- exception will be ex- 

 plained under cannibalism. 



The chief point to be noted, however, under the relation of sex 

 to general bodily form is that, with the advent of maturity, a 

 marked sexual dimorphism appears. The tails of the males under- 

 go a considerable change in form and most noticeably an increase 

 in length. This secondary sexual character seems to be invari- 

 ably acquired simultaneously with the development of enlarged 

 cloacal lips and of the testes. The only exceptions are cases of 

 extremely poor sexual development due to under-nutrition. A 

 few such animals become sexually mature males without the cus- 

 tomary lengthening of the tail. Measurements were made of 

 thirty young males, after metamorphosis but before attainment 

 of sexual maturity, and again, early the following season, after 

 the attainment of sexual maturity. These showed a lengthening 

 of the tail in all but three cases of extremely meager or pseudo- 

 development. That this change was not due to simple growth is 

 shown by the fact that the length of the head and body had in- 

 creased in but two of the most highly fed individuals, while in at 

 least four it had actually decreased, at times by as much as 5 ram. 

 Moreover, measurements of twenty-eight young females from the 

 same lots and kept under same conditions showed an actual 

 shrinkage in entire length in eighteen individuals, this being 

 chiefly due, in most cases, to a shortening of the tail. Of the six 

 individual females that showed growth one only showed a special 

 extension of tail of 2 mm. This lengthening of the tails of the 

 young males with the approach of sexual maturity, though gen- 

 eral, is, like many secondary characters, extremely variable. In 

 four of the above instances it equaled or exceeded 2 cm. ; its aver- 

 age was a little over one centimeter. The increase is entirely by 



212 



