absorbée! il. iliis discovery mav liave an important bearing in 
connection with those flies wliich are known to transmit trypano¬ 
somes to man and the lower animais. A tsetse flv for instance 
Avhich bas become infected with its natural fkigelLate as well as 
witn r. gainbiensc may be in the same starved condition, and 
wotild then perhaps feed oftener, and be liable to transmit T. 
gambiensc to a larger proportion of people than a fly tininfect- 
ed with wild flagellâtes. At présent nothing whatever is known 
regarding the effect wild flagellâtes may hâve on pathogenic try¬ 
panosomes when taken into an insect’s stomach ; my e)bservations 
certainly suggest lhat they may hâve an important bearing on 
these parasites. 
h'rom the fourteenth day onwards the flagelkated forms of Her- 
pelomonas muscac dornesticae begin to tindergo the changes pre- 
paratory to encystment, thev shorten <'ind then divide more than 
once, finallv' rotinding tip either in the flies’ cokm or recttim ; l 
hâve not yet worked ont the time the parasites take to complété 
this cliange. The cycle of events described above only takes place 
in those flies which hâve had liberal feeds of' spleen, liver or 
méat jtiice, and it is probc'ible that the flagellated forms may re¬ 
main as stich for a longer period than I hâve yet been able to 
détermine as the flies only lived tip to the fifteenth day. 1 am at 
présent experimenting witit a larger cage (fig. 3) and hope to 
keep the flies alive for there weeks or more. 
The next important point discovered was that by infecting flies 
in the way described above, tind giving them two subsecpient 
feeds of clean spleen and then only jam and water, the flies con- 
tained cvsts from the sixth day onwaards. If in addition to the jam 
the flies are given fresh horse dting the {tarasites disappear en- 
tirelv from the alimentary tract of the majority of the flies; in one 
stu'h experiment onh' one flv otit of a large ntimber contained 
cvsts on the sixth day. Tliese experiments suggest that jam and 
the jtiice obtained from fresh horse dtmg are not snitable food 
for Herpetomonas muscac domesticae. 
1 hâve pointed above tliat flies may also ingest the preencyst- 
ing (postflagellate) stages of H. muscac domcslicac as these 
forms are commonlv formel in the faeces of the bazaar flies. The 
occurrence of these ])arasites in the stomai'h (mid-giit) of a fly 
is verv apt to lead to the idea, that they represent stages in the 
life c^’cle of another flagellate, for they are AVM'y like some cri- 
