86 
Psyche 
[June 
entotrophan families Campodeidae and Japygidae are the 
preferred prey. In the Pueblo Nuevo forests, campodeids 
are abundant (many were on the limb housing the rain- 
forest colony), and they may well form the principal food 
supply. Honey was ignored by the workers, although 
available in the artificial nest for at least two weeks. 
Despite the rather spectacular development of its man- 
dibles, there does not appear to be anything really unusual 
about this species’ method of catching prey, although it is 
admitted that the workers were never seen in the act of 
hunting uninjured entotrophans, the presumed usual prey. 
When a brood of newly hatched geophilid centipedes was 
placed in the food chamber, the ants rushed them immedi- 
ately, seized them with their mandibles, and shook them 
back and forth with a forward bobbing motion of the 
head. Only one individual was stung, in addition, before 
being carried back to the brood. The Belonopelta, when 
hunting or fighting intruders, do not open their mandibles 
more than is usual for other Ponerini. Also, the mandibles 
are not handled like traps as in other long-jawed groups 
such as the Odontomachini and Dacetini, nor does their 
strike have the stunning effect sometimes observed in these 
groups. My own interpretation is that their peculiar 
shape is a special adaptation for pinning entotrophans, 
which insects are very active and agile, and difficult for 
most ants to hold and sting. 
My Belonopelta were generally very timid, in most in- 
stances fleeing frantically from arthropods not sought as 
prey, including the docile Nasutitermes workers. Their 
mandibles crossed one another at rest as shown in figures 
1 and 2 and were never opened to threaten intruders. 
When the workers transported larvae, they cradled them 
between the concave masticatory borders and avoided using 
the needle-like apical teeth. 
The Belonopelta larvae were very active ; when disturbed 
they thrashed violently back and forth in the manner of 
injured earthworms, but showed no capacity for directed 
locomotion. Insect prey were fed to them in typical poner- 
ine fashion on their “laps”, either entire or cut up into 
large pieces. The cicadellid mentioned above was placed 
entire across the laps of two large larvae lying side by side. 
