THE H^MOFLAGELLATES. 



241 



the writer has not come aci'oss any other observations of 

 multiplication-rosettes being formed by the parasites while in 

 the blood. 



On the other hand, as above indicated, the occurrence of such rosettes 

 has frequently been observed in artificial cultures of different Trjpanosomes. 

 Fig. 29 shows two such clusters in the case of T. lewisi. The larger ones 

 are formed apparently by the successive divisions of the elements of the 

 smaller ones ; concurrently the individuals gradually lose their pyriform 

 shape and become more elongate and fusiform. In this way colonies of 

 hundreds of individuals are formed. 



The origin of these rosettes appears to be by the multiple division of a 

 single form. The radial arrangement and the general shape of the para- 

 sites in a small cluster suggests this (cf. fig. 29 a with fig. 28 v, showing 



Fig. 29. — Multiplication-rosettes of T. lewisi from a culture. 

 (After L. and M.) 



segmentation in the blood). Novy and McNeal, moreover, describe and 

 figure (81) early stages in such multiple division in T. avium and other 

 Avian parasites, which they consider will lead on to the formation of a 

 rosette. Probably, however, " segmentation " is soon replaced by rapid 

 binary division. A noticeable distinction from the multiplication-rosettes 

 of T. lewisi in the blood is that, in most cases described, the clusters of 

 Trypanosomes so formed in cultures have their flagella and kinetonuclei 

 centrally disposed.' It is by no means impossible that this results from the 

 strange medium in which the parasites are ; in so far as this is the case, 

 the arrangement must be considered abnormal. 



' This is not so in T. brucii, which, according to both Smedley and 

 McNeal, preserves the " blood-type," — i. e. the flagella are outwardly 

 directed. 



