130 - CONTROLLING SEX IN BUTTERFLIES. 
set in a large box partly filled with earth, the whole being cov- 
ered with deep blue mosquito-netting. Heat and moisture seemed 
favorable to health and rapid growth. 
On the 25th of June one lot of eggs hatched, on the 10th of 
July they were chrysalides, and on the 18th of the same month the 
butterflies appeared, only requiring twenty-three days for the com- 
plete transformation. On the other hand, I have had this same 
Asterias butterfly eleven months in coming to maturity; some 
larve that hatched in August, 1871, I fed eight weeks, but the 
nights were cool and some days were absolutely cold, when 
the larve would not eat. These chrysalides I preserved during the 
winter, and early in June, 1872, I put them in this same warm 
room in which the larvvæ grew so rapidly, and they were in 
this room some two weeks before the first larvee of this season 
were hatched; and strange as it may appear, some half dozen 
butterflies of this year’s brood came out before these last year’s 
chrysalides produced butterflies. 
Very soon after the last moult, I shut a number of the larve 
away from food, putting them in paper .boxes, from five to ten in 
a box, carefully labelled. If, at the end of two or three days, the 
larvee were still wandering about, I fed them sparingly; in this 
way I did not lose a single specimen in the larva state by shutting 
away from food; a few of the chrysalides died. 
It was with the most intense interest that I watched the coming 
forth of the butterflies, which began to appear in about eight days 
after assuming the chrysalis stage. Thirty-four males came from 
my male boxes, and then a rather small female made its appear- 
ance. Out of seventy-nine specimens that I labelled males, three 
females were produced. On the other hand, those that I fed up, 
keeping them on a good supply of fresh food, I labelled females, 
and placed them in separate boxes. Out of these boxes sixty- 
eight females came and four males. 
There were some boxes that I marked doubtfal, which I do not 
include in the above figures. For instance, I took five larvae that 
were eating vigorously; if let alone they probably would have 
eaten a day or two longer, but I wished to try them in all stages 
of growth, and these were of quite a large size; out of these five, 
four were females. 
Soon after the last moult, I took twenty larvee and shut them 
away from food for twenty-four hours. At the end of that time I 
eos 
