282 TROCEEDINGS OF THE ACADEMY OP 



fouj* da3's the}' began to assume the chrysalis form. After several 

 weeks I removed the usual silken covering, as has always been 

 m\^ custom, from the fact that past experience has shown me that 

 the development proceeds more rapidly, especially v/hen they are 

 kept in a room where the temperature through the cold season of 

 the yesxY falls not short of 10 Fah. I was surprised to find that 

 those which had changed first turned out to be males without ex- 

 ception, while the last batch, consisting of a dozen cocoons, proved 

 with but two exceptions to be females. Here was a problem to 

 solve, and in reflecting upon the subject it occurred to me that 

 possibly Mr. Meehan's lata of nutrition would furnish me with an 

 explanation of it. I conducted a series of experiments during the 

 ensuing summer, to test the truth of my observations. Accord- 

 ingly in August and September I procured besides specimens of 

 the larvce above mentioned, many of the Ceratocampa regalis, 

 Dryocampa imperialism Sphinx quinquemaculata, S. Carolina, 

 Snierinthus excsecata, and others of the Sphinx family. These 

 worms were taken while passing through the period intervening 

 between the last and the preceding changes. In nearlj' ever}' case 

 the w^orms were selected as shortlj' after the next to the last change 

 as was possible. These worms, amounting to three scores, were 

 divided into two equal sets and placed in separate boxes. The 

 one set was deprived of all food; the other was kept provided 

 with fresh, wholesome fare in abundance. The members of the 

 first set, after the expiration of a few days, began to appear ex- 

 ceedingl}' restless, as if looking for some convenient place to hang 

 their cocoons, or to deposit their chr3'salides. Those of the second 

 kept on feeding for a week or ten daj's, and then began to undergo 

 their final transformation. The difference in the appearance of 

 the two sets was quite perceptible. The former were small, and 

 presented the look of animals that had been ill-fed ; the latter 

 were large and plump, and showed evidence of having fed sumptu- 

 ously. After allowing sufficient time to elapse for the chrysalides 

 to form and harden, I began the work of divesting them of their 

 cocoons. 1 found that the chrj^salides of the one, as the greater 

 breadth of the antenna^, which was already mapped out, would 

 seem to indicate, were invariably males, while the other, judging 

 from the narrowness of the antenna^, proved with but two excep- 

 tions to be the opposite sex. The chrj-salides of the Sphinxes j^re- 

 sented a similar state of alfairs. The appearance of the perfected 



