the changes Avhich it goes through are known as metamorphoses. 
Entomologists have generally considered the life of all insects to be 
divisible into four well-marked periods — that of the egg ; the larva 
or caterpillar ; the pupa or chrysalis ; and, finally, the imago or per- 
fect insect. It is true that in some orders, as, for instance, the Coleop- 
tera (beetles), Hymenoptera (bees), Lepidoptera (butterflies), and 
Diptera (flies), the larvae differ much more from the perfect insects 
than is the case in others, as, for instance, the Orthoptera (grasshoppers), 
or Heteroptera (bugs) ; but even in these latter the stages were still 
supposed to be well marked — that of the larva, by the entire absence of 
wings ; tiiat of the pupa, by the possession of rudimentary wings ; 
finally, the perfect insect, by having perfect wings. 
The lecturer then pointed out that, when the habits were alike, 
similar larvae might be met with in very different families of insects. 
Thus, among beetles, the Melolontha (cockchafer), Anobium (death- 
watch), and Chlamys are very similar in their larva state, although they 
belong to perfectly distinct families of beetles, namely, the Melolon- 
thidae, Ptinidae, and Chrysomelidfe. 
The same fact holds good even in larvae belonging to diiferent 
orders of insects. Those larvae which, in the words of Mr. Herbert 
Spencer, are " symmetrically related to the environment," and which 
either are surrounded by their food, or which have it brought to them, 
are fat, legless, fleshy grubs or maggots. Such are almost all the larvas 
of flies. So again, the Hymenopterous larvae are generally of this cha- 
racter ; whether they inhabit other insects, like those of the ichneu- 
mons ; or live inside galls, like those of the Cynipidae ; or are enclosed 
in cells and fed by the perfect insects, like those of the bees ; practi- 
cally any great deviation from that which may be looked upon as the 
normal type, is unnecessary. The larvae of beetles, on the contrary, 
are generally of a very different character. But tliere is one group, 
that of the weevils, which are internal feeders. The grub of a nut- 
weevil feeding inside a nut, is under very similar conditions to those of 
a Cynips-larva in a gall, or an Anthrax-larva living parasitically in a 
bee's cell, and we accordingly find that these larvae, though belonging 
to three different orders of insects, very closely resemble one another. 
To this type belong most Hymenopterous larvae ; but there are two 
exceptional groups, the Tenthredinidae, or sawflies, and the Siricidae. 
Thelarvieof the Tenthredinidae feed, like those of butterflies, on leaves, 
and in the general form of the body, in the possession of three pairs of 
legs and several pairs of abdominal prolegs, they very closely resemble 
ordinary caterpillars, and differ extremely from the ordinary type of 
Hymenopterous larvae. In the same manner the larvae of the Siricidae, 
which are wood-borers, possess thoracic legs, and closely resemble the 
larvae of some wood-boring beetles. 
From these facts it may be concluded that the form of a larva depends 
more on the conditions in which it lives than on the form which it will 
ultimately assume. But this is shown still more clearly in the case of 
Sitaris, a small beetle which is parasitic on a species of solitary bee 
