122 
Hciemoflcigelicit es, etc. 
necessary. A certain percentage of the second generation of leeches 
were also infected with flagellates, and these again were regarded as 
stages of T. inopinatum, yet we can find no proof that this is 
actually the case. In a species of Glossiphonia we found that 
about 80—90 °/o of the embryos from an infected parent had 
Crithidia ; these flagellates were in no way connected with a frog 
trypanosome, nor was it possible to transmit them to healthy frogs. We 
therefore believe that Brumpt has fallen into the error of describing a 
natural Crithidia of H. algira which is transmitted hereditarily. We 
can well understand, that as a certain percentage of embryos of this 
leech are not infected, all their progeny would remain clean, and that 
although they may be fed ever so many times on infected frogs, they 
will themselves never become infected; in our opinion this does not 
indicate that the leech is refractory, but that T. inopinatum does not 
undergo any developmental changes in the leech. In one experiment 
forty uninfected leeches were fed on infected frogs for ten days, and then 
isolated for seven days, after which they were placed on a clean frog 
which later became infected. This experiment in our opinion only 
shows that T. inopinatum may be transmitted after at least seven days 1 ; 
we would like to know if there were any unchanged trypanosomes 
(. T. inopinatum ) in the sheath of the proboscis of the leeches. 
In the case of Helobdella stagnalis Brumpt found trypanoplasmata 
abundant in the proboscis of specimens of this leech; a large percentage 
of the embryos of an infected parent were themselves infected. We are 
left in doubt as to the origin of this trypanoplasm, and would like to 
know whether it is a true Trypanoplasm a or a Crithidia of the leech 
transmitted hereditarily. Brumpt has not published any figures of the 
flagellate. 
Brumpt has examined some hundreds of embryos of Hemiclepsis 
marginata descended from adults infected with various species of 
trypanosomes and trypanoplasmes of fresh water fish, but he was never 
able to demonstrate hereditary transmission. Similar results were 
obtained with Piscicola geometra. Is it possible that the flagellates of 
these two leeches are transmitted in some other way ? In every instance 
Brumpt has failed to exclude a natural flagellate of the leeches he 
was dealing with; surely this is a grave source of error. Now that 
it is known that these flagellates commonly occur in leeches, it is first 
necessary to exclude them before it is possible to trace the development 
of any other parasite which may be ingested by these invertebrates. 
1 This experiment at once recalls Kleine (1909) and Bruce’s recent work. 
