FERTILITY OF CECROPIA EGGS. 247 



all of the eggs that were shrivelled proved infertile and contained 

 only a dried mass of yellow substance. It also usually happened 

 that the smooth, round eggs contained a full-formed insect. 

 There were a few exceptions in each class, however, and external 

 appearance was not relied upon, but all eggs were dissected. 

 Again a number of fertile eggs had openings of various sizes, 

 sometimes too small to free the insect, and at other times per- 

 fectly cut, but the caterpillar it seems felt not inclined to leave 

 the shell. These were in every case counted with those which 

 normally emerged. 



FERTILITY OF EGGS AND DURATION OF MATING. 



The table shows the fertility of the eggs in relation to the time 

 spent in copulo. All those which continued for about the normal 

 length of time, 16 hours or more, behaved naturally and severed 

 of their own accord; the others (3 to 15 hours) were for the most 

 part artificially separated to ascertain the effect of the shortened 

 period upon the fertility. The table is divided into first day's 

 deposit, second day's deposit, etc., after mating. 



An inspection of the columns reveals that there is absolutely 

 no relation between the period of copulation and the fertility of 

 the eggs. 1 We find that the insects which remained in copulo 

 from 3 to 15 hours deposited just as large a per cent, of fertile 

 eggs, and also as large actual numbers, as those which continued 

 from 1 8 to 48 hours. Hence our present material at least 

 would lead us to conclude that the first three hours are quite 

 sufficient, and that a longer period is only a waste of valuable time 

 and vitality. Especially does this seem distinctly disadvan- 

 tageous to the species when we consider the number of eggs which 

 the females produce but can not deposit in their short lives. If 

 females which remain in copulo only 3 hours deposit as large a 

 per cent, of fertile eggs as those which continue for 24 hours or 

 more and afterward die without completely ovipositing, it would 

 seem as though natural selection had missed its mark in not 

 shortening the period of copulation and thereby lengthening the 

 chance for oviposition. 



1 The per cents of fertility were also calculated and tabulated for each individual 

 of the entire lot, but these also showed no relationship whatsoever. Since the 

 data are submitted it was thought superfluous to publish these per cents here. 



