In the course of an earlier paper (Thorpe and Jones 1937), it was shown that the ichneumonid parasite
Nemeritis canescens
(Grav.) can be induced to develop a positive olfactory response to the odour of an abnormal host, the wax moth
Meliphora grisella
(F.), by rearing it artificially upon that host. It was shown that the change in the olfactory response brought about by this process was due, in large part, to conditioning taking place in the adult stage immediately after emergence from the pupa. This was confirmed by the fact that
Nemeritis
, reared on the normal host
Ephestia
, can be made positively responsive to the odour of
Meliphora
merely by placing them immediately on emergence in an apparatus through which is pumped a stream of air which has previously passed over a number of living
Meliphora
larvae. But the conditioning, obtained by treatment of the newly emerged adult in this manner, is not usually as strong as that produced by actually rearing the insects on the abnormal host. From this it was concluded that part at least of the conditioning effect produced by rearing on the wax moth larvae must be the result of the influence of the host acting during the pre-imaginal period. Attempts to get conclusive demonstrations of this in
Nemeritis
were, however, ineffective. It was hoped that it would be possible, having reared the parasite to the pupal stage on the wax moth, to dissect it from its cocoon, which of course may be presumed to be contaminated with the odour of the host. For if pupae, having been thus dissected out and then washed and completely isolated from the odour of the wax moth, still show significant preference for it on emergence, the existence of pre-imaginal conditioning could be considered firmly established. Unfortunately, the high mortality caused when pupae are treated in this way made the experiment impracticable. Many parasites can be removed from their cocoons without suffering harm, but
Nemeritis
appears much more susceptible to injury than most. The object of the present work was to perform a similar experiment on some insect which does not spin a cocoon and in which the puparia can therefore be washed free of all contaminating odoriferous substances. Because of its convenience as a laboratory insect
Drosophila melanogaster
Mg. (
fasciata
Mg.,
ampelophila
Loew.) was chosen for this purpose (a strain being kindly supplied by Dr C. H. Waddington). The two sexes were used indiscriminately since in some preliminary control experiments no significant differences in their behaviour were observed. In a second paper (Thorpe 1938), it was shown that in
Nemeritis
a positive conditioning can be produced, not merely to the odour of a possible host, but to the odour of such biologically abnormal substances as cedar wood oil. It was felt that
Drosophila
also offered a particularly suitable subject for further experiments of this nature in that its rearing on synthetic foodstuff lias long been standardized. Odoriferous substances can easily be mixed with the food, thus exposing the larva to their influence over the whole of its life. The
Drosophila
were reared on Pearl’s S. 101 medium, modified in certain particulars by Mr J. H. Sang, for whose help in the matter I am much indebted. The exact composition of the medium is as follows:
Habitat selection: Olfactory response of Drosophila melanogaster depends on resources
