STUDIRS ON AVIAN H.EMOPROTOZOA. 703 



also Novy and McNeal's figures of so-called multiplication- 

 rosettes on PI. 8) is probably to be regarded, indeed, as indi- 

 cating the tendency of these forms to become attached, when 

 in the natural iusectan-medium. In the culture-medium, how- 

 ever, there is nothing for them to attach themselves to, ex- 

 cepting these commencing clusters of their fellow-individuals. 

 Hence, the probable explanation — in great measure, at any 

 rate — of the clumps or clusters which have their flagella 

 centrally directed, is that they represent the attached phase 

 in the insect. This is of well-known occurrence, both among 

 trypanosomes (cf. Prowazek [I.e.], figs. 53 and 54), and 

 among insectan flagellates (cf. especially Patton [16, PI. 9, 

 fig. 22], where a number of Crithidia sp., in Gerris are 

 clustered ai'ound a food-particle, and again, Swingle [32], 

 who states that a rosette of Crithidia in the sheep-ked, 

 Melophagus, may be formed around a free epithelial cell). 

 In the case of parasites in cultures, when one, two, or tnree 

 individuals have become entangled by their flagella, the inter- 

 locked ends fui-nish doubtless the "nucleus" for the attach- 

 ment of many other parasites, with the result that a large 

 cluster is soon formed. 



An important point brought out decisively by my cultures is 

 that this avian trypanosome does not proceed to form 

 rounded-off, resting phases immediately on passing from the 

 vertebrate host into the cold medium. And further, I may 

 mention, there is not the least indication of any such behaviour 

 in the case of the trypanosome of the little owl when it 

 passes into the stomach of the mosquito. 



Up to the present only one or two accounts of cultural 

 forms of trypanosomes have been published which describe 

 and make any attempt to distinguish between the different 

 types of form and phases developed at different periods in 

 the culture. Of these, the most important for purposes of 

 comparison with my own results is the paper of Novy and 

 McNeal, to which reference has been made. In this connection 

 it must be emphasised that most of the authors' figures of 

 ■cultural forms (and apparently their descriptions also) are 



