244 
C7'itMdia pulicis n. sp. 
may rapidly follow, but the original dividing organisms have separated 
from one another as a rule, so that rosettes are not produced. 
Division of the flagellate form occurs just previous to the formation 
of the post-flagellates, but the number of divisions appears to be less 
than in the other Grithidia I have studied, and as a result, the ovoid 
post-flagellates of C. pulicis are relatively larger. 
The form of the parasite is influenced to some extent by the diet of 
the host. In a starved flea the body cytoplasm of the flagellates is very 
finely granular. A short time after fresh blood reaches these parasites, 
they become larger, more richly granular and division commences. 
Flagellates occurring in the semi-digested “ dark ” blood in the hinder 
part of the alimentary canal of the flea are larger on the average than 
those in the fore part, the food medium apparently being more suited 
to their needs. 
Method of Infection. 
The method of infection of P. irritans by G. pulicis is purely a 
contaminative (casual) one, so far as I can ascertain. The faeces of 
already infected P. irritans contain n\imbers of post-flagellate forms 
of G. pulicis and sometimes also some active flagellates. Other fleas 
may obtain blood near the spot of skin fouled by their neighbours and 
thereby ingest some of the post-flagellates. In other cases, as the 
insects moved about, especially in the confined space of my “ flea cages,” 
fluid dejecta contaminated their bodies and legs. In the removal of 
the offending material, the post-flagellates are brought in contact with 
the mouth parts and thence find their way into the alimentary canal 
of the host. 
I have seen no indications of hereditary infection. Careful dis¬ 
section of both male and female genitalia failed to show any form of 
the parasite therein on examinatioii. Breeding experiments confirmed 
the absence of Grithidia in the eggs. Clean fleas bred clean. Eggs of 
infected fleas hatched apart from the parents also yielded uninfected 
stocks. I have never found G. pulicis in any situation other than 
the alimentary tract and dejecta of P. irritans. 
Some Remarks on Insect Flagellates, especicdly of the Genus Crithidia. 
Earlier in this paper I referred to the fact that bred fleas were used 
during my investigations and that my blood was the medium on 
which they fed. This raises two possibilities. According to the 
