DEVELOPMENT OF CYNIPID HYPERPARASITES 471 
as an ecto-parasite upon the pupa. Here presumably the chitin 
protects the larva during the search for the host. Charips 
is an endoparasite throughout larval life, but certain facts 
suggest that this may be a later adaptation, and that the 
chitinous armour may be a survival of a life-cycle not unlike 
that of Perilampus. 
For instance, the chitin does not now seem to be of vital 
importance to the young larva, since it may either be thrown 
off at hatching and left behind in the embryonic membrane 
or persist for a variable number of days afterwards. Smith (24) 
suggests that the histolysis of the surrounding tissues is the 
stimulus that impels the Perilampus to change its mode 
of hfe and moult. Something of the kind may occur in 
Charips, though in this form metamorphosis of the host 
does not actually take place. The host larvae may be in 
different stages of development at oviposition, and yet those 
younger than the third instar could scarcely contain enough 
food material to enable the Cynipid to reach maturity. It is 
doubtful whether in such a case as that shown in fig. 3, where 
the gut is already displaced before the hyperparasites have left 
the embryonic membrane, the Aphidius can survive. But 
even in ovipositions in third-instar Braconids it would be fatal 
to the Cynipid if the development of the host were arrested 
too soon, for instance before the cocoon was woven. Thus it 
is possible that the chitinized stage is in some sort a resting 
phase, and I now regret that I did not pay more attention to 
this point in the material at my disposal. 
Another point is that Perilampus is endoparasitic only 
in the first instar, whereas Charips lives internally until 
larval development is completed. 
But a parallel may be drawn if the internal habit of the 
latter is a comparatively recent adaptation, and the demolition 
of the host’s remains after emergence is a survival from a time 
when it made its way out of the host at an earlier stage and 
completed development as an ectoparasite. 
The metabolism of Charips presents certain problems. 
The thick chitin must prevent cutaneous transfusion of oxygen 
