170 
Psyche 
[December 
crevices on the tops of enormous boulders twelve to fifteen feet high 
and twenty to thirty feet across. The base of each of these boulders 
was embedded in the gravel and sand of the canyon floor and the 
back face of each was buried in a mass of rubble that formed the 
bottom of a talus slope extending downward from the steep canyon 
wall. It may be seen that the crevices in which clydei was living were 
at least twelve feet above the canyon floor and even further removed 
from the talus slope at the rear of the boulder. The nests appeared 
to be completely isolated from any contact with soil although, since 
it was impossible to determine the extent of the crevices, there was a 
remote chance that they might extend through the boulder to the 
soil at its base or to the talus slope at the rear. The action of the for- 
agers made this seem unlikely, for if there had been any soil connec- 
tion at the bottom or rear of the boulder heavily laden minors would 
scarcely have struggled to the 1 top of the boulder when returning 
to the nest with food. But the character of the nest at Horse Tanks 
definitely ruled out any chance of a connection with the soil. This 
nest was situated on the face of a ledge which overhung the pool or 
“tank” at its base. The ledge was about twenty-five feet high and 
so nearly vertical that it could not be climbed. All that could be 
done was to watch the ants until they went out of sight up the ledge. 
But it was plain that there could be no connection with soil here, for 
the ledge was a part of a basin of solid stone in which the pool lay. 
Thus it seems clear that clydei customarily nests in the crevices of 
large boulders or ledges and not in the soil. 
The writer knows of no other North American species of Phei- 
dole which behaves in this fashion. While several species of Pheidole 
( grallipeSj subdentcita, hyatti, etc.) will sometimes nest in the crevices 
between separated layers of stone, these crevices are always filled 
with soil and the nest passages run into the soil in which the layers 
of stone are buried. Such nests are not isolated from the soil, as are 
those of clydei. This peculiar nesting response of clydei is not easy 
to explain. The writer at first believed that clydei selected nest sites 
that would protect it from flash floods, which are heavy and destruc- 
tive both in Deep Canyon and in the Split Mountain area. There is 
enough truth in the above view to make it dangerous, for the elevated 
position of the clydei nests undoubtedly puts them above flash flood 
levels. But this view fails to take account of the fact that clydei 
might secure equally good flood protection, as do most of the ants 
which live in Deep Canyon, by nesting in the talus slopes above the 
flash flood levels. The protection afforded by nests in rock crevices 
may be of another sort. In several of the areas where clydei occurs 
