296 
Journal of Agricultural Research 
Vol. XXVI, No. 7 
did appear at Gamboa the rain had ceased and the winged adults came 
in ever-increasing numbers from the woods north of the town and drifted 
out over the Canal, where large numbers perished. This flight was still 
in progress when observations were discontinued at 6 p. m. Since there 
was no rain at the time on the Pacific side of the Divide, no termites were 
found swarming at Ancon or Panama City. 
In a region like the Canal Zone and the adjoining parts of the Re¬ 
public of Panama, where there is such a decided difference, even within 
short distances, not only in the quantity of rainfall, but also in the time 
during which it falls, one place being deluged and another remaining 
dry, it is apparent that there will be a considerable range in the time 
that swarming takes place even in the same species, for it seems that 
accumulated rainfall is a factor in determining the time that this event 
occurs. Species like those of Nasutitermes, Amitermes, and Anoplo- 
termes, whose main nests are built in the ground and which require moist¬ 
ure in order to continue their work, must swarm early in the rainy season 
in order to establish themselves thoroughly, for during the dry season, 
from the last part of December to the first or middle of April, practically 
no rain falls over a considerable part of the region and the ground becomes 
hard, dry, and cracked for several feet down. 
The fact that such large numbers of “true” queens have been found 
in the same carton nest leads to the conclusion that in this species 
winged individuals are more or less gregarious even after fertilization. 
Because of the number of queens that occur, the nests of this .species 
are crowded with workers and soldiers and this seems to be an advantage 
in keeping out intruders such as the ants in case the nest or runways are 
damaged. On July 19, 1919, in the Las Sabanas region of Panama the 
carton nest shown in Plate 2, A, was cut open with a machete to 
obtain queens. The outer parts of the stump on which this nest was 
found were inhabited by the termites and the inner parts of it by car¬ 
penter ants, Camponotus abdominalis Fab., subspecies stercorarius 
Forel. 12 The ants and termites, as has been repeatedly observed, were 
living in apparent peace, either unaware of the existence of each other 
or in a state of “armed neutrality.” In opening the termite nest with a 
machete the stump was also split and the termite hordes poured forth 
only to meet the throngs of furious ants. Each apparently blamed the 
other for the catastrophe that befell its nest. The ants rushed at the 
termites with open jaws, closing them and crushing as many as half a 
dozen at a time. But the taste of termite blood was not pleasant and 
the ants soon let them fall. The worker termites grasped their enemies 
by the legs and antennae while the nasuti with their “beaks” borne aloft 
and secreting the white, milky fluid from the tips thereof, rushed at the 
invaders as if to pierce them through. Their favorite point of attack 
was the abdomen, especially the pedicel of the ants, and here as many 
as four nasuti were found attached to a single invader by the fluid they 
had secreted, while half a dozen workers were clinging to its legs and 
antennae. Though this fluid seems to have no effect on the skin of man, 
it apparently paralyzed the ants and numbers of them were seen to curl 
up, lose their hold on the stump, and fall to the ground helpless. If an 
ant did succeed in freeing itself from one lot of termites it was soon 
covered with another and it seemed unable to do anything when these 
got on its back or on its abdomen. The ants soon became discouraged, 
13 Determined by W. M. Mann of the Bureau of Entomology. 
