14 BULLETIN 766, U. S. DEPARTMENT OF AGRICULTURE. 
sides were holes covered with fine copper screen, affording good venti- 
lation. In the center of the front of the tray was a hole 3 inches in 
diameter, in which was fitted a round plug having a 1-inch hole 
through the center for inserting a vial containing foliage; by this ar- 
rangement the stem of the foliage could be kept in water. The experi- 
ment number label, with the number and species of hosts, number of 
both sexes of Compsilura, and the date begun, was also pasted on the 
front. In the right side of the tray was a smaller hole for the purpose 
of placing the parasites in the tray, this being closed with a cork. 
During the summer of 1915 sugar water was used as food, being 
sprayed on the leaves and on the sides of the tray. This was found 
to be unsatisfactory, first, because sugar Water was not heavy enough 
food for the flies and they did not live for any length of time, and, 
second, when this sticky substance was sprayed over the leaves and 
the sides of the tray the flies frequently would become stuck to it. 
During the season of 
1916 honey water in 
the proportion of one 
part honey to three 
of water was used 
very satisfactorily. 
This was fed to the 
flies by the use of an 
atomizer, with which 
it was sprayed on a 
sponge placed in a 
watch glass. In the 
bottom of the tray 
= : : = was kept a small dish 
FIe. 9.—Reproduction tray used in life-history experiments of sand. which was 
on Compsilura concinnata. (Original.) 3 
moistened each day, 
thus keeping the humidity constant. Temperature records were kept 
with a self-recording thermometer placed near the reproduction trays. 
The experiments were conducted in a house which had three screened 
windows on one side and the opposite side built so that about half 
of it was open and well screened. One end was closed and the other 
had a screen door in it. The three sides being thus open, good ven- 
tilation was afforded. The reproduction trays were arranged on two 
sides of this shed on shelves which were constructed of narrow strips 
to facilitate ventilation in the trays. 
During the summer of 1915 difficulty was experienced in securing 
-parasite-free hosts, the only material of this nature that could be 
had in large numbers being the brown-tail moth larve. These were 
reared from the hibernating webs collected the previous fall. A 
large number of gipsy-moth larve were hatched for this purpose, 
i al 
