220 PARASITES OF GIPSY AND BROWN-TAIL MOTHS. 
would have been reared before now from some among the hundreds 
of thousands of brown-tail caterpillars which have been carried 
through their first three or four spring stages in the laboratory. Xone 
having been reared under these circumstances, the only logical con- 
clusion is that they start into activity so early and develop so rapidly 
as to cause the death of the host before they are sufficiently advanced 
to pupate successfully. This is not necessarily the true explanation 
of the failure to rear the species from hibernating brown-tail cater- 
pillars fed in confinement, but it appears to be the best. 
Ordinarily in the summer the larvae do not pass over into the pupa 
of the host, but occasionally they do so. In the late summer and 
fall, when the host caterpillar is of a species which hibernates as a 
pupa, the parasite appears to be aware of that fact in some subtle 
manner, and likewise prepares for hibernation. Its larvae (or what 
are without much doubt its larvae) have several times been found in 
hibernating pupae of several species. The adult has never yet been 
reared from pupae under these circumstances, and the record is on 
that account open to some question. 
The larger part of the Compsilura which were imported from 1906 to 
1908, inclusive, issued from puparia (PI. XX, fig. 1) found free in the 
boxes of brown-tail caterpillars from abroad. A companion species, 
Dexodes nigripes, which is indistinguishable from Compsilura in any 
of its preparatory stages, has also been reared under exactly similar 
circumstances, but curiously enough, if Compsilura was common in 
material from the same locality, Dexodes was apt to be rare, or vice 
versa. Some few were reared from gipsy-moth importations during 
this same period, but not in anything like the numbers which were 
secured from the brown-tail moth material, and it was not considered 
as of particular importance as a gipsy-moth parasite until 1909, 
when it was found to be very common among the tachinid parasites 
secured from shipments of gipsy-moth caterpillars from southern 
France. 
The first colonies of Compsilura were planted in various localities 
within the gipsy-moth infested area in 1906, and in 1907, according 
to the records of the laboratory, a single fly was reared from gipsy- 
moth caterpillars collected in the immediate vicinity of one of these 
colonies. There is some reason to doubt the truth of this record, 
since every attempt at recovery made in 1908 failed. 
In 1907 a much larger colony than any ever liberated before was 
located in the town of Saugus, in the near vicinity of one of those of 
the previous season. In 1908 none was colonized. In 1909 several 
very Large and satisfactory colonies were planted in several places 
within the infested area, and for the first time it was felt that the 
species had been given a fair opportunity to prove its effectiveness 
as an enemy of the gipsy moth and brown-tail moth in America. 
