THE STRUCTURE OF TRYPANOSOMA LEWISI. 769 
various ways, principally with the Romanowsky stain, with 
iron-hematoxylin, and with ‘'wort’s stain. Some of the films 
stained by the Romanowsky method were dried off (Pl. 22, 
figs. 70, 73), but as a rule they were kept wet and passed 
through acetone and xylol into balsam. All films stained with 
iron-hematoxylin or with Twort’s stain were kept wet through- 
out and mounted in balsam. 
After previous fixation with osmic vapour the trypano- 
somes generally show a distinct periplast line (figs. 17, 18, 
75), not to be seen when the sublimate mixture is used 
directly. There is little other difference to be noticed 
between osmic-fixed films and those in which osmic has not 
been used. With either method iron-hematoxylin gives good 
results; T'wort’s stain, however, does not work well after 
osmic fixation, but stains very well after fixation direct in the 
sublimate mixture (figs. 82-84). 
(b) Schaudinn’s fluid (two volumes of saturated solu- 
tion of corrosive sublimate in water, one volume of absolute 
alcohol, with the addition of a few drops of glacial acetic).— 
This fluid was also used with (figs. 25, 24) or without (figs. 
25-28, 76, 85) previous exposure of the film to osmic vapour. 
The films were never dried and were mounted after staining 
in Canada-balsam. The results were on the whole similar 
to those observed with sublimate acetic. A periplast-liue 
can generally be observed, and appears to depend on the 
degree of extraction of the stain, as is well seen after 1ron- 
hematoxylin (figs. 23-38). ‘Twort’s stain gives very good 
results if the fixing fluid be used without previous osmic 
vapour-fixation (fig. 85). 
(c) Mann’s picro-corrosive.— Made up as follows: 
2°5 grm. of corrosive sublimate dissolved in 100. ¢.c. of 
boiling distilled water ; when dissolved, 1 grm. of picric acid 
added ; the mixture allowed to cool, and either used as made 
up, or with addition of 15-20 c.c. of formol, mixed in imme- 
diately before use. Wet films were fixed in the mixture and 
stained in various ways without drying at any time. I could 
perceive very little difference resulting from the presence or 
