474 MAUD D. HAVILAND 
host comes into force, if not at oviposition, at least at an 
early stage in larval life. 
In support of the view that the larva itself may inhibit 
the development of the host is the parallel case of Peri- 
lampus. As the larva is hatched as a free-living form and 
subsequently enters the host, there can be no question of the 
inhibition dating from oviposition. Yet, according to Smith 
(24), ‘The development of the host ... invariably ceases at 
the time of exit of the planidium. Whether or no it is actually 
killed is not evident. In any case decomposition does not 
take place immediately, the host being left in a condition 
somewhat comparable to that of the prey of certain aculeate 
Hymenoptera.’ 
Perilampus differs from Charips in that metamor- 
phosis has taken place before the exit of the planidium ; but 
when the latter begins to live as an ectoparasite upon the 
newly-formed pupa, it is found that the growth of the head 
and appendages, with their setae and pigments, is arrested, 
and development is not completed. 
Nothing resembling phagocytic reaction against the hyper- 
parasite was observed, either as regards the living larva or 
the cast skin, which could sometimes be found unchanged 
among the host’s tissues up to the time of emergence of the 
full-grown Cynipid larva. 
EcCoNoMIC STATUS. 
Charips checks the Aphidius in its destruction of plant- 
lice, and thus, from the economic standpoint, must be con- 
sidered an injurious insect. But throughout its development 
it shares the vulnerability of its host to ectoparasitic Chaleids 
and Proctotrypids, and when secondary parasitization occurs it 
perishes with the Aphidius. From observations made in 
the course of this work it would seem that where the incidence 
of Chalcid and Proctotrypid hyperparasitization is high, the 
chances of Charips larvae attaining maturity are corre- 
spondingly reduced. For instance, if, of ahundred A phidius, 
twenty-five are parasitized by Charips, and_ thirty-two 
