32 W, A. LAMBORN. 



In the first experiment, from 16th May to 14th June, 20 pupae were subjected to 

 five female Mutillids, and from these three male and five female parasites were bred 

 between the 14th July and 3rd September. 



A fresh experiment was commenced on 14th June and finished on 7th July, by 

 which time 250 pupae had been subjected to 20 female Mutillids. Between 24th 

 August and 26th September 36 male and 75 female parasites emerged from these 

 pupae. 



A third experiment was commenced on 7th July and terminated on 4th August^ 

 by which time 200 pupae had been subjected to 24 Mutillids. As a result, 19 male 

 and 35 female parasites were obtained from 20th August to 6th October. 



Up to 6th October, therefore, the 470 pupae subjected to female Mutillids have 

 yielded a total of 173 of the parasites (58 males and 115 females), which gives a 

 percentage of about 36, and a balance of 114 pupae still remains. Some of these 

 may be dead, for they have suffered many vicissitudes, but a few w^hich have been 

 carefully chipped show the Mutilla cocoon within, so that the percentage may be 

 considerably higher. 



An examination of the 9,285 pupa-cases found between 7th April and 2nd June, 

 from the same locality from which the pupae utilised were taken, shows that 353 had 

 been parasitised, a percentage of rather over 3, and in the course of further work in 

 the same neighbourhood, from 22nd August to the 16th October, 7,731 cases have 

 been obtained, Avhich show parasitism by the Mutillids in 487 instances, or 6'3 per 

 cent. In neither set of figures has the question of hyper-parasites, which are rela- 

 tively scarce, been taken into consideration. 



In connection with the figures quoted, it is perhaps well to emphasise the fact that 

 the material composing the Mutilla cocoon is unaffected even by long soaking in 

 water, and is so tough as to be proof, as experiment has shown, against the attack of 

 ants, the insects most likely to destroy it. 



The length of time occupied by the Mutilla in its development within the tsetse 

 pupa seems to vary so very considerably as to lead one to suspect that emergence 

 must be determined by factors other than developmental ones, possibly climatic 

 changes. In the case of the Mutilla bred from a brevipalpis pupa development 

 occupied 12 weeks three days, whereas Mutillids are even now coming out from 

 pupae found in mid-May, fully 17 weeks ago. 



With the advance of the dry season there has been a remarkable increase in the 

 numbers of Mutilla glossinae, in fact of parasites of all kinds, that have emerged from 

 pupae in my possession, so much so that the Mutillids now outnumber the tsetses. 

 Out of 762 new pupae obtained from natural breeding places between 22nd August 

 and 16th October, 97 of the flies, 45 males and 52 females, have so far emerged, as 

 compared with 131 Mutillids, 62 males and 69 females. Surprising figures were 

 obtained with wild pupae, all of which were taken close to a water-hole, which has 

 dried up during the last four months, from breeding places which were all carefully 

 searched before for pupae in May. Under shelter of one particular fallen log in a 

 coppice were found on 24th and 25th August no less than 40 pupae (90 new ones having 

 already been removed from the same place in May), and these pupae with others found 

 thereabouts, amounting in all to 85, were utilised for the purpose of ascertaining the 



