340 
They occurred either within the stomach, in the stomachal tissues, 
or in the body cavity of the mosquito. A central differentiated area 
could be detected in them by examination in fresh specimens; near 
the area were more or less numerous granules which often exhibited 
lively Brownian movement. After some time the parasites became 
much stumpier (8/i by 16/i), and were later almost spherical. One 
parasite, however, remained unchanged and active for twenty-four 
hours. 
Further search showed that the parasite occurred in freshly- 
caught and freshly-hatched mosquitoes, and in the larvae from which 
all our experimental mosquitoes were raised, and it was quickly shown 
to be a stage in the life cycle of a coccidium. It was most interesting 
that practically every larva from one pool was infected with this 
coccidium, while none of the larvae and adults (anopheline and 
cuhcine) from a pool only a few yards distant were affected. So far 
as was observed the parasite did not occasion any excessive mortality 
amongst the mosquitoes infected. This seems difficult to understand- 
the mosquitoes were probably not watched long enough—since the 
tissues of many of them were fairly riddled with coccidium cysts. 
Hie water of the heavily-infected pool was centrifugalised and 
examined. Coccidium cysts were not recognised, various infusoria 
and a clump of herpetomonas-like flagellates were alone seen. 
So soon as it was certain that the vermicule-like “sporozoite” hac 
no connection with the trypanosomes ingested, the study of this 
coccidium was discontinued. It is, therefore, not possible to say 
w ether the forms observed represent stages in more than one 
^ ° r no ' ^ seems probable that but one species of coccidium 
s present. We describe the forms observed and indicate the 
position they seem to occupy in the life cycle of the parasite. The 
mobile sporozoites first seen become free through the rupture of a 
sporoblast containing eight sporozoites. The number of sporozoites 
m each sporoblast is almost invariable. Fig. 15 is unusual in that 
nuc eus o two of the sporozoites has divided, and in one instance 
sion o t le cytoplasm seems to have commenced. As a rule the 
ong .ameters of the sporozoites are parallel. These are, therefore, 
gu arly arranged like the segments of an orange (corps en bardlet). 
Single sporoblasts surrounded by a definite cyst wall may occur, 
y are more usually seen in groups of four, seven, or even as many 
