48 THE CANADIAN ENTOMOLOGIST. 



is small, and both the upper and lower surfaces are perfectly accessible, 

 the absence of leaves is but little missed. But, on the other hand, where 

 these substitutes do not exist, the chrysalis has been known to appear 

 without the customary covering. 



This fact seems to point to the conclusion that the cocoon is only a 

 subsequent acquirement, which did not primarily exist. In climates 

 where rains are of common occurrence, as protecting envelopes, they are 

 indispensable. A chrysalis will endure a very low temperature while 

 comparatively passive, with perfect impunity ; but cannot endure excessive 

 moisture without destruction. 



Of the exact time which the chrysalis requires to develop into the 

 imago, I can only say that it depends upon thermometric conditions; were 

 the several larval transformations undergone in early summer, while the 

 mercury is standing at 92 degrees, and were the chrysalis stage then 

 assumed, it is probable that the moth would appear in about two weeks. 

 We reason from analogy. Actias lii/ia requires but a single day less than 

 a fortnight to pass from the condition of pupa to that of imago, and surely 

 cecropia could scarcely surpass this period. Cocoons that have been 

 taken into the house in August, and kept close to a hot stove, have 

 developed in January; while those which have been left out doors, seldom 

 change before the middle of May. As far as I have been able to ascer- 

 tain, this species is single-brooded. 



Before bringing this sketch to a close, there are a few particulars which 

 I shall touch upon, that came under my immediate notice during the 

 season that has passed. It is a mistaken impression that caterpillars of 

 particular species confine their feeding to certain plant-species, exclusively. 

 During several years past, I have closely looked into this subject, and my 

 experience has been otherwise. My observations upon cecropia have been 

 both interesting and remarkable. In the neighborhood of Germantown, 

 the leaves of the common red currant, constitute its favorite food. During 

 the first and second stages, by which I mean the intervals before and after 

 the first moulting, it entirely restricts its feeding thereto. But after the 

 second moulting, it readily accustoms itself to Ribes nigrwn,R. grossidaria, 

 Primus cerasus, P. vulgaris, Rosa blanda and Spircea coryvibosa. A little 

 later, I have tried numerous larvte upon Wistaria sinensis, Philadelphus 

 inodorus, Syringa vulgaris and Prunus serotina, with remarkable success. 

 Subsequent to the last moulting, several caterpillars were induced to feed 

 upon SympJioricarpiis racemosus. Some cocoons which were produced by 



