D. L. Mackinnon 
247 
makes its diet almost exclusively off algae and diatoms,—bacteria were 
relatively scarce, and I never saw any trichomonads. 
I could detect no differences between the trichomastix from one 
genus and that from another. 
In all, 100 larvae were examined ; of these all but 14 were infected 
in a greater or less degree, and of the latter, 6 belonged to the genus 
Mystacides. 
I examined 13 pupae and 20 adult caddis-flies, but never found any 
trace of the flagellate. 
The parasites occupy a very definite position in the gut of the larva. 
The upper portion of the intestine, from which the Malpighian tubules 
come off, is considerably wider than those regions of the gut immediately 
above and below it. There is therefore a tendency for debris to collect 
at this point, and the intestine is sometimes almost blocked by a rich 
growth of bacteria. Chief among these is a species of Streptothrix, its 
filaments forming dense, tangled masses; a large Bacillus also occurs 
frequently 1 . If the gut be dissected out, and examined under a ^ 
objective, the flagellates can easily be seen through the semi-transparent 
walls, moving about among the thick bacterial growth in the upper 
intestine. They also occur, though in much smaller numbers, in the 
lower part of the intestine, and in the rectum. I have never seen them 
in the mid-gut. 
In order to study the living parasite, I teased out the intestine in a 
small drop of normal salt-solution, which I examined under an oil- 
immersion (Zeiss y^"), after first sealing down the coverslip with wax. 
The parasites are very active at first after removal from the host, but 
soon become sluggish in their movements, and die in an hour or less: in 
this respect they seem more sensitive than Trichomastix lacertae, and 
T. batrachorum, which Prowazek and Dobell describe as living under 
these conditions for a long time. Of the intra vitam stains employed, 
neutral red proved the most useful. Schewiakoff’s method of fixation, 
with osmic acid vapour and subsequent examination in soda solutiou, 
was found very useful in determining the number and point of origin of 
the flagella. 
In making permanent preparations, I tried all the ordinary fixatives 
and stains, but the results given by Heidenhain’s iron-haematoxylin 
after fixation with sublimate-alcohol (Schaudinn’s formula) were so 
much the most satisfactory that I soon trusted to this method alone. 
1 Leger (1902) describes these, or very similar, bacteria from the intestine of the larva 
of Chironomus. 
16—2 
