HYMENOPTERA. 119 
larva, dead bodies of small quadrupeds and birds, &c., constitute 
their food. They feed the larve with their mouths, transport them 
in fine weather to the external superficies of the hill, in order that they 
may receive additional warmth, and take them down again on the ap- 
proach of night or bad weather, defend them from their enemies, and 
look to their preservation with the greatest fidelity, particularly 
when the hill is disturbed. They pay equal attention to the nymphs, 
some of which are enclosed in a cocoon, and the others naked; they 
tear open the envelope of the former when the moment of their ulti- 
mate metamorphosis has arrived. 
I have observed neuters in various ant-hills, remarkable for a 
head much larger than common, and for the unusual fewness of 
their number. M. Dupont de Nemours, without being a naturalist, 
had also previously noticed this difference(1). M. de la Cordaire, 
whom I have already mentioned, has given me a neuter allied to the 
alta cephalotes of Fabricius, and assures me that individuals of this 
kind were the defenders of their community, and apparently fulfilled 
the functions of Captains in their excursions, at which time they 
marched along the sides of the main body. 
The name of eggs is vulgarly appHed to the larve and nymphs; 
those of the F. rufa are eaten by young Pheasants. The neuters 
‘prevent the individuals with newly acquired wings from issuing 
forth until the proper moment has arrived, which is always deter- 
mined by the heat of the atmosphere. They then open a passage 
for them and let them go. 
Most ant-hills are wholly composed of individuals of the same 
species. ~Nature, however, has deviated from this plan with respect 
to the F. roussatre or Amazon-ant, and that which I have called the 
sanguinea. Their neuters, by open violence, procure auxiliaries of 
their own caste but of different species, which I have designated by 
the names of noir-cendrée and mineuse. When the heat of the day 
begins to lessen, and exactly at the same hour, at least for several 
»days, the Amazons or Legionnaires quit their nest, advance in a solid 
column, more or less numerous or according to the extent of the 
population, and march upon the Ant-hill they wish to attack. They 
soon penetrate into it notwithstanding the opposition of the inhabi- 
tants, seize the larve and nymphs of the neuters peculiar to the in- 
vaded community, and trapsport them in the same warlike order to 
their own domicil, where they are attended to in common with the 
posterity of their conquerors, by other neuters of their own species 
(1) See his Recherches sur les Fourmis Indigénes. 
