THE AMERICAN SILK WORM. 149 
month than in July and August, the larve did not grow 
so rapidly, and the moulting did not take place so regu- 
larly. The first moulting took place on the fourteenth 
day, the second the twenty-third day, the third the thirty- 
sixth day; on the first of November, or fifty-six days 
after their birth, they had not accomplished the fourth 
moulting. I could not continue the experiment, as I left 
’ for Europe the second of November; but they had frozen 
several times, and the leaves were very hard, in fact I do 
not believe that the second brood would have come to ma- 
turity. I do not see that it would be of any advantage to 
obtain two broods, as the moths do not all come out of 
the cocoon at the same time, but sometimes there are two 
months between the first and the last; so the process of 
rearing can go on permanently = summer, which is equal 
to having two broods. 
Cion can be retarded in hati out by being put 
in a very cold room—an ice-house, for instance; in this 
way they can be made to hatch another year, or nearly 
twenty-one months after they have been in the cocoon. 
In fact, the time of their appearance can be put back for 
an indefinite period, as life is nearly suspended. Reau- 
mur states, that, at the time he was writing, he had in his 
-cellar pupæ which had been there for five years, which 
_Were still living. I have myself kept pupe of sphingide, 
or hawkmoths, for three years in my cellar. At the time 
I went to Europe, they were still living, but on my re- 
‘turn I found that the rats had eaten them. 
