486 JOURNAL OF ECONOMIC ENTOMOLOGY [Vol. 10 



have not been able to find out whether the mature parasite deposits 

 its eggs cutaneously into the caterpillar, or into the chrysalis, or 

 whether its ova are introduced into the alimentary tract of the larva 

 with its food, as I have found the parasitic larva only in the chrysalids. " 

 To determine this point on the parasitization, large numbers of 

 the caterpillars were collected and bred out. Each season when 

 the nests were cut down from the trees several large ones 

 were secured and hung up in the screened house already mentioned 

 and closely observed for signs of parasites. When secured the cater- 

 pillars were nearly always mature, which would have afforded ample 

 opportunity for any eggs of the parasites to have been deposited on 

 the skin, or to have been eaten, if that was the manner in which they 

 were parasitized. As soon as pupation began cotton sheeting was 

 tacked around the sides of the house over the screening to prevent any 

 flies from depositing their eggs or larvae through the screen onto any 

 of the chrysalids that were attached to the screening. Out of approx- 

 imately ten thousand caterpillars that were collected to be bred out 

 in this manner during different seasons, not a single parasite was found 

 either in the caterpillar stage or in the chrysalids that had transformed 

 inside the house, although all the dead caterpillars and chrysalids that 

 died of fungus or from other causes were carefully dissected and exam- 

 ined. This alone would show that if any were parasitized in the 

 caterpillar stage the percentage must be very small. Aside from this, 

 the fact that the caterpillars remain hidden in the nests during the 

 daytime for the greater part of their lives would help to exclude any 

 chance of being parasitized during this stage. 



As proof that the chrysalis is the form attacked, as many of the 

 chrysalids as could be found were collected each season and a large 

 percentage were found to be infested with either dipterous or hymen- 

 opterous larvse. In 1914, during September and October, 106 chrys- 

 alids that had pupated out-of-doors in different places were collected 

 and placed in separate jars to breed out. Judging from the time that 

 some of the adults subsequently emerged and from the fact that many 

 had been allowed to remain outdoors for several days after being found, 

 before they were taken inside, the chrysalids had probably passed from 

 6 to 9 days out-of-doors in the chrysalis stage. Out of this number, 

 20 were collected from cocoanut trees. Adult butterflies emerged 

 from 14 of this lot, 3 died of fungus, and 3 were destroyed by dip- 

 terous parasites. 



Sixty-nine were collected from outbuildings in the rear of the Lab- 

 oratory containing rabbit hutches and small animal cages. These 

 buildings are about ten feet distant from a cocoanut tree and seem to 

 be a favorite place for pupation. Forty-five adults emerged from 



