138 E. A. :MTNr'HIX AND If. M. WOODCOCK. 



membrane is usually twisted in a spiral fashion round the 

 body by tLe voluntary contortion of the protoplasm, this 

 beino- in all probability effected by the contraction of 

 myonenies. Minchin (13) has recently published figures 

 clearly showing myoneines in T. percse andT. granulosum, 

 and we have no doubt they are present in other fish-trypano- 

 sonies, though we have not had the good fortune to see 

 them in T. raite. In wet films the parasites have retained 

 their twisted position. In "dry" smears, on the other 

 hand, the body becomes untwisted, and, at the same time, 

 passively or mechanically coiled up in one plane, by the 

 mere fact of the attached membrane being longer than the 

 body is. 



We do not think this different behaviour on '*di-y " smears 

 is to be explained l)y a flatteuing-out process due to actual 

 drying. In tlie first place, in uur procedure, the slides are 

 removed from the osniic-acid tube and placed in absolute 

 alcohol before the moisture dries off from the greater part, 

 at all events, of the slide ; it is only along tiie edges that 

 drying sometimes occurs. And after the smear has been 

 hardened in alcohol little or no alteration, we consider, takes 

 place in the form of the parasites, even though the smear is 

 allowed to dry off ultimately. In fact, as Minchin has 

 already shown in his account of the technique in connection 

 with T. lewisi (14), this method is probably the best for 

 the general form and size of the parasites. Secondly, now 

 and again where the body of the parasite really appears to 

 be somewhat flattened out due to an actual drying at first, 

 this C-form is not shown (cf . tig. 30 of T. trigise). Indeed, 

 this process of untwisting and coiling would seem to require 

 the presence of a film of moisture for its accomplishment. 

 The following explanation appears to us the most probable. 

 In fixation by the "wet" method, both death and fixation 

 are practically instantaneous. In fixation by osmic vapour, 

 on the other hand, death probabl}' occurs appreciably before 

 fixation. In the twisted condition, during life, tlie fiagellar 

 border of the membrane is probably to a certain extent in 



