438 THE AMERICAN NATURALIST. (Vor. XXXV. 
corner of the nest a number of Myrmica workers had formed 
a circle about a few of their small larvae which they were 
cleansing and feeding. A Leptothorax soon found its way to 
this cluster and stepped from the back of one ant to that of 
another, lavishing a shampoo on each in turn and apparently 
filling its crop with the liquid contributions thus solicited. 
This and numerous very similar observations, which could 
be made at any time on removing the cover of the nest, prove 
conclusively that the Leptothorax workers demand and obtain 
their food from the Myrmica workers. The method of solicit- 
ing food, however, differs from that of any other myrme- 
coxenous animals known to me. These animals usually request 
food by tapping the ant with their antennae (many myrme- 
cophilous beetles), or stroking its face with their fore feet 
(Atemeles), but none of these guests are so unconventional as 
to mount the backs and scratch the heads of their hosts for 
the purpose of inducing the latter to regurgitate. Even the 
slave-holding Polyergus and the social parasite Anergates 
demand and receive food after the manner of other ants. The 
Leptothorax workers are so persistent in their peculiar atten- 
tions to the Myrmicas that I have come to doubt whether the 
little guest ants ever really feed themselves. Once only was 
a Leptothorax seen to approach the dish of syrup, lap up 4 
very little of it hastily, and then return to the nest. This 
happened before the ants had definitely settled under the pane 
of glass. After that only the Myrmica workers visited the 
manger, and the Leptothorax usually waylaid them as soon as 
they had entered the nest. During my first visit to the four 
natural Myrmica-Leptothorax nests I found the latter species 
loitering in the outer galleries just under the stone. I am con- 
vinced that they prefer this situation in order to be on hand 
the very moment a food-laden Myrmica enters the nest. The 
Leptothorax workers often walked on the lower surface of the 
roof-pane, although they had to turn over to mount the passing 
Myrmicas. They must do this also in the natural nests, for 
the specimens taken August 1 were, as above stated, crawling 
on the lower surface of the stones. 
Although the feeding of the guest ants commonly proceeds 
