1. Experiments were carried out with the object of devising means of maintaining large numbers of Glossina in the laboratory, especially for experimental work in trypanosomiasis.

2. Most of the experiments were done with the local species, G. swynnertoni, but maintenance of G. morsitans, G. pallidipes and G. austeni was also attempted.

3. Maintenance at a constant temperature of 26° C. was a complete failure, chiefly owing to the impossibility of maintaining an adequate humidity with the apparatus used.

4. Maintenance in the laboratory (with higher temperatures and consequently lower r.h. for most of the twenty-four hours) gave very poor results and inadequate length of life.

5. Longevity of flies was much improved by keeping them in an open-sided thatched hut in cages over water, and mean lives of up to 110 days in a cage of ten flies were achieved—even over the most severe part of the dry season.

6. Flies were fed in one of two ways: on guinea-pigs in a ‘Geigy’ feeding-rack or on sheep in a crush by fixing the cages round the flanks and abdomen of the sheep with an elastic band.

7. Earlier comparisons of the two methods of feeding either were inconclusive or favoured guinea-pig feeding. After prolonged maintenance the G. swynnertoni fed on sheep gave much the better results. G. pallidipes, from the beginning of maintenance, did much better on sheep-feeding.

8. The population of G. swynnertoni fed on sheep has increased progressively up to the sixth and seventh laboratory-bred generation; the guinea-pig-fed population showed an early increase but has since declined sharply.

9. Fertilization of females in the laboratory has been easily achieved with G. swynnertoni and G. austeni. G. morsitans has shown some failures, and with G. pallidipes great difficulty was encountered.

10. Records of pupal weights and weight-loss have been kept and methods of keeping pupae studied.

11. Results indicated that total weight-loss of pupae was dependent chiefly on humidity and but little, if at all, on temperature; the pupae were therefore finally kept at laboratory temperatures at constant 90% r.h.

12. In early generations of G. swynnertoni and G. morsitans there was significant negative skewness of the distribution curves of initial weights of pupae, indicating an excess of small or ‘runt’ pupae. Flies emerging from these under-weight pupae were eliminated from the maintained population; the lower limits of initial pupal weight, below which the emergent flies were excluded, were 22 mg. for G. swynnertoni and 24 mg. for G. morsitans.

13. In general, there has been a decrease in the mean weight of pupae from generation to generation, except that the course of the decrease has been very different in the two differently fed populations. In the guinea-pig-fed population weights were high in the early generations and later dropped sharply; in the sheep-fed population weights were rather lower at first but have decreased only slightly and then, in the latest generations, increased a little.

14. There has been an increase in the total weight-loss of pupae from pupation to emergence in later generations.

15. The mean interlarval period has been consistently longer than the known period for Glossina at the temperatures experienced, indicating a proportion of abortions up to as much as 50%.

16. Total weight-loss of pupae was found not to be a constant amount as has been stated elsewhere but to be in proportion to pupal size.

17. A consistent sex-difference in total pupal weight-loss was found in the pupae of G. swynnertoni examined, the males losing 1·1–1·6% more of their initial weight. Data are given to show that this difference is fully established within the early period of rapid weight-loss and only increased a little by the greater loss in the males in the last few days before emergence.

18. The early results of maintenance of G. austeni have shown remarkable longevity and very good reproduction-rate, both exceeding the records of other species maintained.

19. It is suggested that G. austeni may prove to be a valuable laboratory vector of trypanosomes in view of these promising results in a climate very different from that of its natural habitat.