32 
Psyche 
[Mar. 
indicative that isoamyl salicylate possesses a strong selec- 
tive attraction for this one fly, Melanophora roralis 
(Linn.) ; but the termination of the experiments then pre- 
vented further confirmatory observations. 
With the later-acquired knowledge of the identity of the 
insect and after reference to Professor Brues’ account of 
its parasitic status, it seemed desirable to give roralis 
closer attention through the succeeding summer. How- 
ever, that endeavor encountered unexpected difficulties, 
for not only was 1947 generally recognized as a year of 
unusual insect scarcity, but in May and June of that year 
2700 acres of Marthas Vineyard, including the area under 
observation, had been sprayed from the air with DDT, 
and the oniscid host of roralis was not present in num- 
bers approaching those of the preceding year. Only by 
persistent search and daily observation did some features 
of the economy of the insect become apparent. 
Upon my arrival on Marthas Vineyard, June 22, roralis 
was promptly detected as a “window-pane” fly. There- 
after, through the summer, five bait-traps yielded one 
hundred and six examples of roralis; many others were 
recorded from the windows, screens and walls of the 
house, indoors and out. These adults occurred over two 
well-defined periods — the first extending from (at least) 
late June to mid- July, then after nearly a month’s in- 
terval (actually by August 10) the flies commenced to 
appear again and were present throughout the remainder 
of the month and as late as September 5, when oppor- 
tunity for observation terminated. These dates seem to 
indicate that in New England roralis has at least two 
annual broods. In England (Thompson) the winter is 
passed as second-stage larvae and no summer brood ap- 
pears. 
Search for the early stages of roralis was continued 
throughout the summer, more successfully under the loose 
bark of fallen trees, where sowbugs often congregate. 
In early July a few empty and broken puparia, each partly 
enclosed in the dry fragments of a sowbug, were found. 
On July 23 two living larvae and one puparium were found, 
each in a body of a recently deceased sowbug. Mean- 
