TRYPANOSOMES IN TSETSE-FLIES AND OTHER DIPTERA. 223 
but without result; it is a pity the experiment was not tried 
of contaminating their food with the faeces. Hence the results 
to hand of observation and experiment, though they are very 
incomplete, indicate that the trypanosomes of the pathogenic 
group, suchas T. bructi, belong to stage 4 of my hypothetical 
phylogeny. ‘Nevertheless, if encystation occurs in one species 
of trypanosome it may occur in others. 
Contaminative infection implies the infection of the verte- 
brate by way of the digestive tract. It is surprising how 
often the occurrence of accidental infections of this kind 
has been noted (see Laveran and Mesnil [28]) and yet 
how seldom it has been the object of direct experiment. 
Thus Bruce ([8], p. 46, Experiment 225) records the case of 
a dog which became infected with nagana after eating a piece 
of coagulated blood from the heart of a diseased heifer. It 
is usual to explain away such infections by supposing that an 
animal which becomes infected by way of the digestive tract 
must have had somewhere a lesion through which the 
trypanosomes penetrated, but this is pure assumption, and 
from the analogy of Dourine-infection it is quite as feasible 
to suppose that infection from the digestive tract can take 
place through the mucous membrane. If the trypanosomes 
can resist the digestive juices of insects, they may also resist 
those of vertebrates, and in this connection I may refer to the 
discovery by Léger (25) of an intestinal Trypanoplasma 
in the fish Box boops. All these facts seem to me to render 
perfectly possible and even probable the contaminative 
infection of the vertebrate, by way of the digestive tract, 
not merely as an exceptional occurrence, but as the normal 
course in those cases where, as in T. grayi, intestinal cysts 
are formed by the trypanosome. I may remark that if we 
start our phylogenetic deductions from Léger’s hypothesis, it 
is not easy to explain the occurrence of encystation in 1. 
e@rayi, unless we deny to it a vertebrate host altogether—a 
difficult position, it seems to me, in view of the habits of 
the fly. 
The foregoing arguments may seem to many too specula- 
VOL. 52, PART 2,—NEW SERIES. 17 
