290 
Flagellates in Fleas 
The first contained relatively few parasites, and these mostly en¬ 
cysting or encysted ; the other two showed a much richer infection, 
with numerous flagellate forms in the upper reaches of the hind-gut. 
Patton has divided the life-cycle of such organisms into three stages, 
pre-flagellate, flagellate, and post-flagellate. I propose to follow the 
same order in my description. 
(1) Pre-flagellates. I found no such stage, all the parasites in the 
upper part of the hind-gut being already adult flagellates or forms 
beginning to encyst. This points to the possibility that the early stages 
previous to flagellation may be gone through in the larva, and until the 
larvae and pupae are examined, the account of the life-cycle must remain 
incomplete. 
(2) Flagellates. By far the greater proportion of the parasites 
were in this stage. There were a few free forms with intact flagellum 
(Plate III, Fig. 1), but the majority were grouped in radiating masses, 
attached to one another in the centre of the rosette by the reduced 
flagella (Plate III, Fig. 3). The form of the adult flagellate is extremely 
long and attenuated (21g x 2p, without reckoning the flagellum), taper¬ 
ing posteriorly to a mere filament. The nucleus is usually in the anterior 
half of the body: it is oval to round in form (about 1/x in diameter) 
and stains purplish-red throughout, with a few minute darker granules. 
The blepharoplast lies anterior to the nucleus, usually at about a 
distance of 1 to 2p, but may be closely apposed to its anterior border. 
The form is a blunt rod, but may be dumb-bell-like. The flagellum 
is complete only in a few instances; in favourable specimens it can 
be traced back almost to the blepharoplast as a faintly reddish streak: 
in one instance it could be seen to originate in a minute, red-staining 
granule, slightly in front of the blepharoplast. In most cases the 
parasite forms one of a group, or has attached itself to the gut-wall 
preparatory to encystment, and the flagellum has become very short, 
or is reduced to a pink-staining mass of apparently viscous matter 
(Plate III, Fig. 2 a, b, and c). The cytoplasm stains delicate lavender to 
purplish. The portion behind the nucleus is usually highly vacuolated, 
and sometimes contains a few small chromatin-like granules. Anterior 
to the nucleus it stains more deeply, taking on a pink colour in the 
neighbourhood of the flagellum. Between the nucleus and blepharo¬ 
plast there is sometimes a vacuole. 
The great variability in size of the flagellates is very striking. 
Alongside the larger forms occur small forms, measuring only about 
7/a x \g, or even less. This may possibly be accounted for by the fact that 
