110 
Psyche 
[March 
a sharp ledge which appears in side-view as a small tooth. Petiole 
relatively longer and lower, its peak rounded, not transversely mar- 
ginate. 
Specimens examined: 1 soldier and 2 workers from Brazil, Ama- 
zonas State, km 20 of Humaita-Porto Velho Road, April 10, 1975, 
Virgilio Pereira da Silva, Daniel Z. Araujo & Aldo J. P. Dillon leg. 
(WWK n. 11926). 
Variation. The worker specimens diagnosed above disagree from 
Mann’s description of the type in having six mandibular teeth in- 
stead of five, and in lacking a longitudinal carina on frontal area. 
The first difference is probably due to an oversight by Mann, 
since the basal mandibular tooth is very small and offset, and hidden 
under the clypeus when the mandibles are firmly closed. At any rate, 
there is no doubt about the conspecificity between the types of bran- 
neri and the present specimens. 
Discussion. The striking and curious dimorphism shown in head 
shape between soldiers and workers (see Figs. 1 and 3), the latter 
possessing a stalked necklike occiput — to my knowledge a unique 
feature for a Camponotus — the shape of the thorax in which the 
metanotum is dorsally exposed and projecting as a transverse welt 
both in soldiers and in workers, the transversely impressed and 
saddle-shaped basal face of propodeum, separate branneri from 
all other species-groups and/or subgenera of the Neotropical region. 
It is hard to point out any closer relationship to any one of the 
other groups, and even more difficult to derive branneri from any 
one of them. 
The soldier head, which is elongate, parallel-sided, and more 
heavily sculptured dorsally in front, together with the rather smooth 
integument of the remainder of the body, reminds one of the more 
orthodox members of subgenus Pseudo colob op sis, but the latter 
lack the standing hairs on scapes and legs, and their thorax is much 
more compact. 
The impressed dorsal profile of thorax of branneri resembles 
superficially that of Myrmosphincta, but in the latter group the 
soldier head is not elongate-rectangular, the clypeus anteriorly not 
impressed, and the metanotum, when exposed dorsally, is deeply 
sunk in between the mesonotum and the propodeum, if not reduced 
to a mere transverse sulcus. 
One could try to derive branneri from the larger, shinier and more 
slender members of the Tanae my r me x-group, especially from those 
