128 Psyche [August 
and high-posted greenhouse, where host and parasite were free to 
follow the natural course of their instincts. They indicate how 
difficult or rather impossible it is for a parasite ever quite to exter- 
minate its host, as some individuals, perhaps in more exposed posi- 
tions, will be parasitized again and again, whereas others, probably 
less exposed, may escape destruction altogether. 
An interesting problem that deserves more attention than we have 
had time to bestow upon it is the fate of the supernumerary larve. 
The few observations already made upon this point are, however, 
instructive. In most cases the supernumerary larve were appar- 
ently already dead, being sometimes somewhat disintegrated though 
bearing no marks of violence. As in all the cases where dead super- 
numerary larve were found there was still an abundance of tissues 
and liquids in the host for food, we cannot suppose that they were 
starved, nor, as there were no marks of violence found on them, can 
we suppose that they were killed in an active combat with a larger 
and older larva. We may be unwilling to believe that the more ad- 
vanced and stronger larve secrete or excrete some fluid or material 
into the body of the host which eventually destroys their younger 
or weaker brothers and sisters, but this view seems forced upon us. 
This gains the more credence when we stop to consider that by some 
subtle influence the weak and often insignificant parasite frequently 
causes important pathological changes in the body tissues of its 
more powerful host. We have noticed this phenomenon not only 
in case of parasitized aphids, but also in caterpillars, notably in 
brown-tail caterpillars parasitized by Meteorus. The change we 
have in mind is the breaking up of the flaky fat-bodies into small, 
more or less globular bodies which are unattached and float freely in 
the body fluids. When in this condition the fatty tissues may be 
more accessible to the parasite as food. 
In no case were the dead larve found surrounded and being de- 
stroyed by the phagocytes of the host, as we have lately observed 
in studies on the larve of Limnerium, the results of which we hope 
to publish shortly. 
DESCRIPTION OF THE LARVAL STAGES OF PRAON AND LIPOLEXIS. 
1. Praon simulans Prov. 
First Stace Larva: Body 14-segmented, somewhat tapering behind; last 
segment with a dorsal-median, cylindrical appendage nearly as long as the 
