882 THE AMERICAN NATURALIST. [Vor. XXXV. 
of the worker ant while she is still a larva. This is proved by 
the fact that two of the large macroérgates are callows, one of 
them still very soft and pale yellow, the other — again the one 
represented in the figure — with harder integument, but with- 
out the deep coloration of the mature workers. Such huge 
parasites could scarcely have made their appearance in ants so 
recently escaped from their pupa. But even if there had been 
no callows among the macroérgates, the truth of the above 
statement would still be patent, both because the macroérgates 
were all infested while none of the normal workers were found 
fo contain parasites, and because the stature of an ant is, of 
course, fixed in the pupal stage and cannot be subsequently 
increased to the dimensions exhibited in the cases under 
consideration. 
It is evident, furthermore, that the macroérgatic stature, 
which is very apparent not only in the distention of the abdo- 
men but also in the greater dimensions of the head, thorax, 
petiole, antennze, and legs, can have its origin only in an unusu- 
ally large amount of food consumed during the growth period 
of larval life! Now, as I have shown in former papers? differ- 
ent species of ants employ very different methods of feeding 
their larva. Species of Camponotus, Formica, Lasius, and 
Myrmica feed their larvae with liquid food regurgitated from 
their crops, and possibly also with the secretion of the salivary 
glands. Other species, however, like the Ponerinz and some 
Myrmicinze (Aphzenogaster, Pogonomyrmex, Tomognathus, and 
some species of Pheidole), feed their larvae with comminuted 
insects. Unfortunately I have not been able to observe the 
method of feeding in Ph. commutata, but it is safe to say that 
it must conform to one or both of these methods. If the larva 
are fed by regurgitation, we must suppose that the parasitized 
1 The opposite condition, że., a small amount of food consumed during larval 
life, results in what may be called microérgatic forms. Such are the firstborn 
workers of all incipient ant colonies. These forms are, of course, perfectly 
normal products of underfeeding, whereas the macroérgates of Pheidole are pre 
ducts of overfeeding induced by a pathological condition. a 
2 Wheeler, W. M. A Study of Some Texan Ponerinæ, Biol. Bull., vol. a 
(1900), No. 1, pp. 1-31, Figs. r-10; and The Habits of Ponera and 5 
tomma, Biol. Bull., vol. ii (1900), No. 2, pp. 43-69, Figs. 1-5- 
