C. Strickland and N. H. Swellengrkbkl 
445 
similarly fed on a series of animals with negative results for about four weeks. 
They were then put on another rat and left on it and this became infected. 
This mechanical transmission by the bite is probably due to the 
inoculation of the blood forms from the mouth parts of the flea, but 
there must be even here a close adaptation of the parasites to their 
environment, for this mechanism is ineffective to transmit T. brucei 
under similar conditions. 
Our experiments seem to indicate that the mechanism of infection 
is not the bite of the flea for ‘late’ transmission, but that it is so for 
‘ mechanical ’ transmission. 
(d) The relation between the develojmiental cycle of T. lewisi in the 
flea and infection by the flea. It stands to reason that the bodies of 
potentially infective fleas must contain T. lewisi in some form or other, 
and consequently that the flagellates should be found in the insects 
upon microscopic examination. It should also be possible to find some 
forms of the tryiianosomes in fleas during the latent period prior to 
their becoming infective in cases where distant transmission takes 
place. We have examined a large number of fleas with the object of 
discovering potentially infective parasites which they might contain, 
with the following results; 
I. Having removed 75 fleas (C. fasciatiis) from breeding-box ‘ A ’ 
in which an infected rat had been kept for some weeks, they were fed on 
a series of ‘ clean ’ rats during a period of 14 days. None of the clean rats 
became infected through the agency of the fleas. All of these fleas 
were dissected and examined during a period lasting for 14 days after they 
had been removed from the vicinity of the infected rat. In only two of 
the 75 fleas were flagellates encountered. These two fleas had not been 
in contact with the infected rat for seven and nine days. The flagellates 
were small in numbers, occurred in the hiudgut, and were of the type 
described on p. 374. 
The examination of the fleas’ salivary glands, ovaries, fat-bodies, 
Malpighian tubes and coelomic fluid gave negative results. The object 
of first feeding the fleas on ‘ clean ’ rats was of course to get some clue as 
to the presence of infective parasites, or by discovering the mechanism 
of infection to deduce the whereabouts of the parasite in the insect. 
II. Another batch of fleas to the number of 239 were removed 
from breeding-box ‘ B ’ in which an infected rat had been confined for 
about four weeks. The fleas were divided into two lots (a) and (b). 
The fleas in lot (a), 20 in number, were dissected and hypothetical 
infective forms were looked for in all of their organs with the result 
29—2 
