THE STRUCTURE OF TRYPANOSOMA LEWISI. PRAT; 
(figs. 10, 11), stained with iron-hematoxylin, and which 
results also from other stains, such as methyl-green, Twort’s 
stain, etc. The resemblance between preparation a, when 
fully extracted, and preparation c, extends also to other 
points. The flagelluim is fine and delicate in structure, the 
blepharoplast minute and scarcely visible, and the periplast- 
line stands out sharply on the concave sides of the body. 
But there is one remarkable point which puzzled me greatly, 
and which I took great pains to assure myself was real and 
not due to errors in representation, namely, that in pre- 
paration a (figs. 64-69), the trypanosomes are constantly 
smaller as a whole than in preparation b (figs. 62, 63). This 
is difficult to explain. It may be that the preparation J has 
dried slightly during the process of fixation, and so resisted 
the tendency to shrinkage, which I believe must always be 
reckoned with in preparations mounted without drying in 
Canada-balsam. But this is evidently not the whole of the 
explanation, for the fully extracted trypanosomes in prepara- 
tion a (figs. 68, 69) are smaller than those in ¢ (figs. 10, 11), 
stained with iron-hematoxylin. It seems to me quite possible 
that when a considerable quantity of stain deposited in the 
body of the trypanosome is dissolved out, the body may 
shrink to some extent as the substance of the stain is 
removed. I do not know how else to explain the distinctly 
smaller size of the trypanosomes in question. 
There is one further point relating to the nucleus which I 
find very difficult to explain, namely, the frequent appearance, 
in trypanosomes stained with Giemsa’s stain in the ordinary 
way, of a lighter spot at or near the centre (figs. 61, 70, 80, 
81, etc.). From a comparison with preparations in which the 
stain is more or less extracted, this clear spot appears to 
correspond to a space or chromatin-free area close beside 
the slightly excentric karyosome; it does not appear, how- 
ever, to be always present. 
I pass to the consideration of the action of the Romanowsky 
stain on the flagellum and periplast. It is well known that 
both these structures stain red with the stain. Schaudinn 
VOL. 53, PART 4.—NEW SERIES. Bs) 
