TRYPANOSOMES IN TSETSE-FLIES AND OTHER DIPTERA. 205 
known with regard to malaria rise at once to the mind. ‘I'he 
researches of Ross, Grassi, and others have demonstrated that 
in malaria the mosquito is a true host in which the parasite 
goes through a complicated developmental cycle, at the end 
of which, and not before, the mosquito is able to infect a fresh 
vertebrate host with the parasite. A mode of infection of 
this kind may be termed conveniently a cyclical type of 
infection, and, since it is effected by the mosquito inoculating 
healthy subjects with the parasites, we may further characterise 
it as the inoculative cyclical type. It is an obvious 
suggestion, from the analogy of malaria, that trypanosomes 
may also undergo a cycle of development in their invertebrate 
hosts, a suggestion that received, apparently, concrete proof 
in the well-known investigations of Schaudinn (40) upon 
Trypanosoma noctux. But it is becoming increasingly 
evident, I think I may say, that Schaudinn’s statements with 
regard to Trypanosoma noctue must be regarded with 
scepticism until they have received confirmation, in view of 
the many possible sources of error from mixed infections which 
his material presented. Nevertheless, Schaudinn’s results 
have been regarded by many as conclusive proofs that 
trypanosomes pass through a developmental cycle in their 
invertebrate hosts. I am conscious myself of having gone to 
Uganda to investigate this question with a distinct bias in 
my mind, believing that the infection of sleeping sickness 
would prove to be of a similar type to that known in the case 
of the malarial parasite, and stated by Schaudinn to occur in 
T. noctuz—namely, the cyclical inoculative type. 
On the other hand, many experimental results, so far 
obtained, indicate that the mode of transmission of sleeping 
sickness, and perhaps of other trypanosome infections also, 
is not of the type of malaria, but is in many cases, at least, 
a direct one—that is to say, that the trypanosome does not 
go through a developmental cycle in the invertebrate host, 
but is inoculated mechanically by the proboscis of the blood- 
sucking intermediary. Thus Brumpt, in 1903 ([7] p. 1497), 
expressed the view that the role of the tsetse-fly could not 
