1987] Windsor — Tortoise beetle, Acromis sparsa 139 
LARVAL DISAPPEARANCE (IND./DAY) 
Fig. 3. Comparison of the disappearance rates (number of individuals disap- 
peared per day of observation) for larval groups at the “forest” (solid bar) and “field” 
sites. 
failed to develop (Table 3). Thirteen pupal masses (48%) produced 
at least some parasitoids. Numerically, tachinid flies eclosing from 
nine groups were roughly twice as important as chalcid wasps. The 
number of parasitoids reared could under estimate their impor- 
tance since some pupal masses were collected soon after they had 
formed. However, I suspect that most parasitoids infect larvae and 
are carried to the pupation site. At least one other larval parasitoid 
that is worth noting is carried to the pupation site, although it was 
not reared from the 1978 sample. Several pupal groups collected in 
the early wet season of 1986 contained larvae of the entomophagous 
moth, Schacodontia sp. (Pyralidae), an important predator on the 
gregarious larvae of the cassidine beetle, Polychalma multicava 
Latreille (pers. obs). 
Sixty-five larvae (17%) in 14 pupal masses of the 1978 sample 
failed to develop— a loss roughly the same as that to parasitoids. 
Again, it is unclear why development in these individuals does not 
proceed. In this case there is the possibility that some individuals do 
not feed sufficiently as larvae before moving en masse with their 
siblings to the pupation site. If there is significant variation in feed- 
ing within groups, then there could be conflict about the timing of 
the descent to the pupation site. Of course, larvae may wither at the 
pupation site because of parasites, fungi, etc. 
