1. Membranes prepared from chicken skin provide a suitable medium through which Aëdes aegypti females may be induced to gorge.
2. Under suitable conditions the proportion of female A. aegypti which will gorge through membranes, though more variable than when a living chick is offered, is great enough for experimental purposes.
3. It is shown that the gorging reaction in A. aegypti is provoked by a heat gradient between the environment and the food-limiting membrane.
4. The feeding reactions of A. aegypti towards whole blood, fractions of blood, and other sub-stances have been studied. It was found that (a) whole blood, and red corpuscles in saline when ingested through membranes go directly into the stomach which becomes fully distended; (b) haemoglobin in plasma or distilled water is ingested to a lesser degree than whole blood or red corpuscles in saline, and plasma alone is rarely ingested, but all these pass to the stomach; (c) sweet solutions containing glucose or honey are seldom imbibed through membranes and pass to the stomach or diverticula, but only the diverticula are fully distended.
When offered as open drops (a) blood is seldom ingested, but if ingested passes to the stomach; (b) haemoglobin in plasma or water, or plasma alone, are very rarely ingested, but pass mainly to the stomach; (c) sweet solutions containing honey or glucose, or mixtures of blood and honey are readily ingested and pass mainly to the diverticula which become fully distended, though traces may be found in the stomach.
5. Aëdes aegypti may be infected with Plasmodium gallinaceum by allowing them to gorge on drawn infected chicken blood through a membrane. Infection rates comparable to those obtained when the mosquitoes are fed directly on living chickens may be obtained by this method.
6. If infected mosquitoes are allowed to gorge upon uninfected blood through membranes they eject viable sporozoites into the blood. When young chicks are injected intravenously with blood so infected, infections are produced which in period of incubation and intensity are comparable with those resulting from the bites of infected mosquitoes.
7. The ejection of sporozoites through membranes in this manner provides a ready means of obtaining sporozoites free from glandular tissue.
8. Sporozoites collected by this method will be suitable for in vitro experiments upon the action of drugs on sporozoites, and also as a source of material for studying in tissue cultures the developmental stages of the malaria parasite arising directly from the sporozoite.
We wish to thank Prof. D. Keilin, F.R.S., and Dr P. Tate for helpful criticism and advice during the course of the work.