188 
Psyche 
[Vol. 95 
on each somite; (3) 0.069-0.088 mm long, with slightly enlarged 
flattened tip, on ventrolateral surfaces of AI-AV1II. Cranium thin; 
subpyriform in anterior view; a third longer than broad. Antenna 
minute; with 2 sensilla at end of a long ridge. No head hairs. 
Labrum a large thin flap, a third broader than long, widest ven- 
trally; with 4 long slender sharp-pointed teeth on each lateral sur- 
face. Mandible turned laterally; apical half of medial surface heavily 
sclerotized; ending in a sharp-pointed apical tooth; anterior surface 
bearing about 4 long slender sharp-pointed teeth; with a wide blade 
bearing 4 long slender sharp-pointed teeth directed laterally. Max- 
illa (only partly visible) with a short base and a large lobate palp 
bearing 5 ( 1 apical and 4 subapical) sensilla; galea not seen. Labium 
a very thin flap, trapezoidal in anterior view, narrowed dorsally; 
palp represented by 5 sensilla on each ventrolateral corner; opening 
of sericteries a narrow slit on the ventral surface. (Material studied; 
5 larvae from Kanagawa Pref., Japan, courtesy of Keiichi Masuko.) 
Apologia 
Of the 800 species of ant larvae we have studied, the leptanilline 
larvae are undoubtedly the most difficult to process. They are min- 
ute: 1 .2- 1 .7 mm long. They are very slender: they must be punctured 
in several places with a minute needle; these minute openings retard 
the transfer of the processing liquids. If KOH does completely dis- 
solve the internal tissues, the insoluble residue cannot be forced out 
without damaging the larval integument. Even when an integument 
is entirely cleaned the stain may never reach the head. After a 
stained integument is on a slide in balsam, it must be moved into the 
desired position for drawing. If the consistency of the balsam is not 
exactly right, the delicate integument will tear. 
Another difficulty: we have had so few specimens (3, 3 and 5) that 
we dared not experiment. 
Once a larva was stained and stably mounted we encountered 
difficulties of interpretation. The most exasperating was determin- 
ing the limits of the prothorax (of this we were never certain). No 
orthodox insect larva or adult should have spiracles on the 
prothorax. But some insects do: the mesothoracic spiracles migrate 
forward during development. But the spiracles of Leptanilla are 
difficult to find because of their minute size. Furthermore Masuko 
has found by SEM photography that spiracles may be vestigial or 
