1937 ] 
Habits of Strumigenys 
103 
Strumigenys during the attack. With these points in mind 
I selected Brachymyrmex heeri var. depilis as most suitable 
although under natural conditions it may be doubted if the 
two species often come in contact. For my purposes the 
choice was a most satisfactory one for each of the thirteen 
Brachymyrmex workers introduced into the nest was killed 
by the Strumigenys although some of them managed to 
avoid death for a number of hours. The method of attack 
employed by the Strumigenys was remarkably constant, so 
much so that I feel it unnecessary to cite individual cases 
unless there is some point of special interest. 
The sequence of events was as follows : 
On admitting the Brachymyrmex to the nest it began a 
series of exploratory investigations during which it visited 
the various nest chambers. As the Strumigenys preferred 
to stay in one of the narrow passages connecting two of the 
chambers it frequently happened that the Brachymyrmex 
actually ran over the quiescent Strumigenys workers. The 
latter roused at once and opened the mandibles but it seldom 
happened that any of them struck at the Brachymyrmex at 
that time. It may be added that, until it was attacked, the 
Brachymyrmex showed little fear of the Strumigenys. After 
becoming aware of the presence of the Brachymyrmex the 
Strumigenys workers would begin to move slowly about 
the nest. It is hard to depict the extreme deliberation of 
their movements. They proceed literally a step at a time 
and this gives to their actions an air of stealth which is prob- 
ably spurious since they never move rapidly under any cir- 
cumstances. Eventually one or more of the Strumigenys 
workers would approach within striking distance of the 
Brachymyrmex. If the latter were quiet the attacker would 
locate its position by cautious explorations with the antenna 
before it struck. Not infrequently, however, the Brachy- 
myrmex blundered into the open jaws of the Strumigenys 
which were promptly snapped shut as the trigger hairs were 
touched. The closure of the mandibles is extremely rapid 
and is followed by a distinct backward jerk of the body of 
the attacker. These two movements seem to be the only 
rapid motions of which the Strumigenys worker is capable. 
The effect of the attack on the Brachymyrmex is very 
marked. It would at once begin an elaborate series of clean- 
