THE CANADIAN ENTOMOLOGIST. 



145 



preserved some examples at every stage in alcohol, and so reduced the 

 number, but there was no death among the larvae which were allowed to 

 proceed, and I obtained at last seven perfect butterflies, i g , 6 $ . They 

 were also of very large size, equal to any taken in the field. It is evi- 

 dent, therefore, that freezing did not injure the larvae, but on the contrary, 

 invigorated them, enabling them to pass their successive stages rapidly. 



Comparing the length of these stages with larvae bred in 1873-4. 



Iced larvae, 1880. 



Time from removal to 1st moult 8 to 18 days. 



1st to 2nd " 8 to 12 



" 2nd to 3rd " ...4 to 9 



3rd to 4th " 5 to 8 



" 4th to 5th " 4 to 8 



" 5th to chrysalis 9 to 12 



" chrysalis to butterfly 16 to 20 



Total period 54 to 87 



Larvae kept in cool room, as related. 



Time from removal to 1st moult ... .44 days and upwards. 



" " 1st to 2nd " 17 " 



" 2nd to 3rd " II " " 



" " 3rd to 4th " .... 12 " 



" " 4th to 5th " 14 " 



" to chrysalis 12 " 



" in " 24 " 



From removal from cool room to imago. 134 " and upwards. 

 From 1 st moult to imago, 90 days. 



I have no doubt that by freezing any species of larvae which hyber- 

 nate, they may successfully be carried to imago — such as Argynnis, 

 Melitaa, Colias, Apatura, Satyrids, Hesperians, etc. Probably it would 

 be better not to remove them from the ice until spring has fairly set in, 

 instead of rousing them prematurely, as I did in case of Cybele. 



Comparing the coloration of the butterflies from the iced larvae with 

 examples taken this season in the field, I see no difference. 



