1979] 
Petr alia & Vinson — Anatomy of Ant Larvae 
391 
to brace or anchor the food while feeding, because of the close 
proximity of this boss to the face of the mouthparts, although this is 
not described in the literature. 
The most efficient method for holding solid food, within the Myr- 
micinae, may be in larvae of Pogonomyrmex barbatus. The segments 
in the posteroventral feeding region (especially abdominal segments 
3-5) bear spinules and strongly denticulate hairs, both pointing to the 
center of each segment. The attenuated thorax allows larvae enough 
dexterity to feed easily from this region. 
This attenuation is even more pronounced in the Ponerinae. The 
larvae of Ponerinae described here have a long neck which allows 
them to reach solid food placed on the wide, flattened posteroventral 
body region which Wheeler and Wheeler (1976) refer to as a food 
“platter.” W. M. Wheeler (1918) describes the feeding behavior of 
Odontomachus: “These larvae are placed by the ants on their broad 
backs, and their heads and necks are folded over onto the concave 
ventral surface, which serves as a table or trough on which food is 
placed by the workers.” This “food platter” is visible in O. clarus and 
Pachycondyla villosa where a row of tubercles demarcates this area 
laterally. It is better developed in P. villosa in which a circle of 
spinules on each segment probably holds food more efficiently than 
the posteriorly pointing spinules in O. clarus. The arrangement of 
spinules in P. villosa may be comparable to that on the posteroventral 
feeding region of P. barbatus (Myrmicinae). 
The feeding region on larvae of the ponerine Leptogenys elongata 
may be even more specialized by the development of a unique cuticu- 
lar process on the ventral surface of the 3rd abdominal segment. The 
cuticular ridges in this structure seem to be formed from a fusion of 
the small, blunt papillae or spinules, as is visible in the peripheral 
parts of the structure. Thus, these ridges are probably homologous to 
the rows of spinules common on ant larvae. We speculate that this 
structure may form a “food tray,” efficient in holding solid food while 
the larva feeds upon it. 3 Recently, John Mirenda and the authors 
have made preliminary observations on larval feeding in live colonies 
of L. elongata. 4 Adult workers place larvae head first into the par- 
3 Dr. G. C. Wheeler and J. Wheeler suggest that this structure might function as a 
stridulatory or food-holding structure or both (pers. comm.). 
4 Collected and identified by Dr. J. Mirenda, from College Station (Brazos Co.), 
TX (March 3, 1980). 
