252 RESEARCH IN PROTOZOOLOGY 



flagellates of each species and an ample series of cross-infection 

 experiments. Whenever possible pure cultures should be made 

 from each species of flea, and these cultures fed to the other 

 species. Drbohlav (1925) has already paved the v^ay for a start in 

 this direction by working out the technique both for cultivating the 

 flagellates and for handling the fleas. The experimental vv^ork should 

 be correlated with careful cytological studies on the flagellates. 

 Other groups can be investigated with the same general problem 

 in mind; e.g., the aquatic hemiptera. Much still remains to be 

 done in the diptera. Wenyon (1926, pp. 1404-1414) has prepared 

 a list of the known invertebrate hosts of the trypanosomid^ 

 which can be consulted for suggestions as to the groups to be un- 

 dertaken for the studies on host-specificity. One would infer from 

 Drbohlav's experiments that cross-infection experiments of hosts 

 considerably removed phylogenetically would give only negative 

 results, but these results should not deter any worker from further 

 attempts of this kind. 



The converse of the problem just discussed is the specificity of 

 the flagellates present in any one host. The writer (1923) has 

 previously pointed out that more than one morphological type of 

 Herpetomonas may be found in muscoid flies. Herpetomonas 

 muscce-domesticce, the most numerous type, has the large parabasal 

 body, often heart-shaped when dividing, prominent ''marginal 

 granules" and conspicuous basal granules. Another type, slightly 

 shorter, has a smaller parabasal body, no marginal granules, and 

 no apparent basal granules. A third shorter and broader type has 

 still another type of parabasal body. There are probably other 

 types which have been overlooked. These different types are prob- 

 ably of specific rank. It may be that Patton's Rhynchoidomonas is 

 a valid genus. It is not unlikely that many of the muscoid flies will 

 be found to have three or four species of flagellate parasites. 

 Critical cytological and biometric studies will solve the problem. It 

 would be extremely desirable to study cultures as well as insect 

 infections. In this connection it should be pointed out that Chatton 

 and his co-workers have identified about five species of trypano- 

 SOMID^ from Drosophila confusa. 



The problem of specificity of herpetomonads has been made to 

 appear even more complicated by the work of Noguchi (1926). 

 He applied, for the first time, extensive bacteriological technique, 

 i.e., agglutination, complement fixation, and carbohydrate-fermen- 



