764. E. A. MINCHIN. 
trypanosomes in preparations made in this way, but they were 
so shrunk and deformed as to be scarcely recognisable. 
When preparations that have been mounted in Canada- 
balsam without being dried at any time are compared with 
those that have been dried off at some stage of the process, 
either before fixation, or after fixation with osmic vapour, or 
after staining, it is found that the trypanosomes in the never- 
dried preparations are constantly slightly smaller than in the 
dried-off preparations. Comparison with the standard shows 
that the dried-off trypanosomes have lost very little in bulk, 
while the never-dried trypanosomes have lost considerably. 
I obtained invariably the same results with the trypanosomes 
and trypanoplasms of fishes; the never-dried specimens 
mounted in Canada-balsam were always a size smaller than 
those in the dried-off films. This result proves, in my opinion, 
that the changes of medium which are incidental to the pro- 
cesses of dehydration with alcohol, clearing with xylol or 
other media, and mounting in Canada-balsam, have the effect 
of diminishing the size of the body. So longasitis kept ina 
semi-fluid plastic condition the body is susceptible to changes 
of medium, and can be shrunken or swollen by them. If, 
however, all fluid be once removed from the body by evapora- 
tion, it appears to obtain a rigidity of texture which resists 
further deformation, although it may be altered in form to a 
greater or less extent by the process of drying itself. 
After these general remarks I proceed to describe in detail 
the effects produced on Trypanosoma lewisi by various 
fixing and staining methods, using as the standard of com- 
parison the preparations fixed simply with osmic acid vapour 
and examined in the fluid blood, as described above. 
(1) Fixatives. 
The fixatives I have used may be grouped conveniently 
under the following heads: (1) Osmic acid vapgpr, followed 
simply by alcohol; (2) mixtures containing osmic acid, (3) 
mixtures in which corrosive sublimate is the principal in- 
