114 SOCIAL HABITATIONS. 
escape. I find that on an average twenty small insects 
are thus found in proportion to one of the larger kind. 
Nothing is easier than the rearing of insects from this 
as well as other galls, but to decide upon the species 
which make them is by no means so easy a task as 
appears on the surface. Hven should the experimenter 
find the right species of insect in the gauze bag, he has to 
go through the wearisome task of searching through the 
family of Cynipide, and identifying the species—a pro- 
cess which every entomologist is rather apt to postpone 
until the visionary period when he shall have leisure. 
But it is very probable that the required insect does 
not make its appearance at all, and that the little 
hymenoptera which make their way out of the cells, or 
are found dead within them, are not the rightful occupants 
of the galls. For the Cynipide are as liable to parasites 
as other insects, and it frequently happens that from a 
single many-chambered gall will issue insects that sadly 
puzzle an amateur, as they seem to belong to at least 
two distinct species. The very gall which has just been 
described affords a good example of this fact, for in some 
of the chambers are specimens of the true Cynips rose, 
and in others are insects which belong to another family, 
the Ichneumonide, which, as the reader may remember, 
are parasites upon other insects. They have evidently 
introduced their eggs into the cells occupied by the 
larvee of Cynips rose, so that the larvee which have been 
hatched from these eggs have fed upon the legitimate 
occupants, and come to maturity in the cells that were 
designed for others. 
Insects of totally different orders sometimes make their 
appearance. When I began to take to pieces the gall 
which has been described, I was rather surprised to find 
among the long hairs an empty cocoon of the Galleria 
