GLOSSINA INVESTIGATIONS IN NYASALAND. 



31 



of the thorax of a tsetse imago, and further evidence of a like character was obtained 

 more recently by finding, on opening another pupa, a fully fed Mutillid larva of the 

 usual Aculeate type (which has since formed a cocoon) with the harder parts, head, 

 proboscis and femora, of a tsetse imago at one end. 



The period of development at which the pupa is parasitised may afford some 

 explanation of the considerable variation in the size of the Mutillids, those which as 

 larvae have had to feed on the imago of a fly so advanced in development as to have 

 tough indigestible parts yielding undersized parasites. Some confirmation of the 

 supposition has been obtained by subjecting G. brevipalpis pupae, which are at least 

 three times the size of those of morsilans, to a female Mutilla. A pupa was seen to 

 be attacked on 2nd July and a male Mutilla, a finer specimen than those obtained 

 from morsitans pupae, emerged on 27th September. This is the more interesting in 

 that I have not been able to obtain evidence that in the ordinary course of nature 

 this Mutilla attacks the pupae of brevipalpis, for an examination of 737 empty cases 

 showed that three only had been parasitised, exhibiting the small round holes that 

 seem to indicate the emergence of large Chalcids, and no Mutillid cocoons were 

 found. Moreover, the situation of the breeding grounds in the depths of thick gloomy 

 bush is not propitious to the activities of the Mutillids, which, like most of their 

 kind, exhibit the maximum activity in the hottest, brightest situations. 



The greater part of the experimental breeding work has necessarily been carried 

 out by means of pupae found in natural breeding places, for though, thanks to 

 Dr. Hearsey, the Principal Medical Officer, the apparatus left by the Royal Society's 

 Commission was placed at my disposal, I was still unable, from lack of a sufficiency 

 of jars and owing to difficulties incidental to work in a tent, to raise a sufficiency of 

 bred pupae. The results are therefore open to the criticism that fallacy may have 

 crept in owing to the use of pupae of unknown history, but apart from the fact that 

 Mutillids have been seen repeatedly to oviposit in the pupae, the high percentage that 

 have emerged, contrasting with the low figures obtained by an examination of the 

 pupa-cases found, does at all events indicate the probability that the insects have 

 been bred on something approaching a large scale. I am now taking measures to 

 obtain a more ample supply of bred pupae for the purpose of estimating the fecundity 

 of the parasite. 



Three large breeding experiments have been conducted. The following table gives 

 details as to the emergence of Mutillids in cases in which parasitism of the pupae was 

 actually seen to be in progress, the pupae being marked and subsequently removed. 



