1915] Cocoons Among Ants 339 
yet no one has ever seen a female Diacamma. In excavating 
the nests of australe, therefore, I scrutinized the ants very 
closely in the hope of finding the unknown female, but in vain. 
Though I searched dozens of nests, I saw nothing resembling 
a winged or dedlated queen or even a worker with conspicuously 
enlarged gaster. I found plenty of larve and pupz and in some 
of the nests during late October a number of males. These are 
bright, reddish yellow, with conspicuously long antenne, and 
quite unlike the bronzy black workers. (Fig. 3). As I failed 
to find any differentiated queen and as all the pupz were of the 
same size, I feel confident that in Diacamma the egg-laying 
function must be usurped by one or more fertile workers during 
the breeding season. 
What impressed me most, while I was excavating the nests 
of australe, was the very dark brown or black color and tough 
consistency of the cocoons. I then remembered that I had 
received very similar cocoons with specimens of D. scalpatrum 
from India and of D. cyaniventre from the Philippines. Further 
study of the nests of australe soon gave a clue to the meaning 
of this unusually deep pigmentation of the pupal envelope. 
On the outer surface of two of the craters, which happened to 
be in shade at the time, I found heaps of cocoons fully exposed 
to the light. At first I supposed that they were the cocoons of 
dead or hatched pup, but on approaching the nests the wary 
workers at once carried them into the nest. Then on examining 
nests exposed to the bright sunlight, I found the cocoons in 
the superficial chambers so near the large entrance that they 
could be distinctly seen through it from the outside. Thus 
while the ants are careful not to place their cocoons in the direct 
rays of the sun they nevertheless expose them to diffuse day- 
light. -The deeply pigmented cocoon evidently enables the 
pupa to utilize the heat rays while effectively protecting it 
from the light and ultra-violet rays. That ants are particu- 
larly sensitive to these latter rays has been proved by the 
experiments of Lubbock (1882), Forel (1886), Forel and Dufour 
(1902) and Miss Fielde (1904). 
It occurred to me that the australe larve might spin a white 
cocoon like other ants and subsequently smear or saturate it 
with the black meconium, or intestinal contents, which all ant 
larve void just before pupating. But an examination of the 
cocoons showed this supposition to be erroneous, for the silken 
