NEW TYPE OF PROTISTAN PAEASITE. 
443 
form thick half-hoops or i-ings, partially encircling the body 
(figs. 18, 19). In large individuals they may occur together 
with conspicuous granules in the peripheral zone (fig. 25), 
but as a rule in the smaller forms, when a prominent 
chromatinic mass is present, the peripheral zone appears to 
contain very little chromatin. 
Division. — Whenever the crescents have occurred in 
numbers Ave have found division proceeding actively. Divi- 
sion of the parasites always 'takes place by means of eqmil 
binary fission. AYe have never seen the slightest indication 
of unequal fission, or of anything in the nature of budding. 
So far as we have been able to ascertain, binary fission 
appears to be the only form of multiplication in S elenomastix. 
The division always takes place in a plane at right angles to 
the long axis of the body, i. e. it is transverse to it. Division 
does not stand, apparently, in any definite relation to the 
size of the parasite ; that is to say, not only large individuals 
divide, but intermediate-sized ones and also quite small forms. 
Neither does the condition in which the chromatinic substance 
is present U;ppear to determine fission, for individuals can be 
found undergoing division in which the chromatin is prac- 
tically in any of the states described above (cf. figs, given 
of dividing forms). 
In the great majority of cases, though not by aii}^ means 
always, the fission is initiated by the splitting of the flagellum 
along the greater part of its length. This is shoAvn clearly 
in figs. 20, 23. There is no question of this appearance being 
merely a fraying-out of the flagellum into fibrils, such as was 
referred to above. For one thing, the instances figured (and 
others observed) are ou wet-fixed films, in Avhich the fraying- 
out is never found. Again, when the flagellum shows a 
frayed-out appearance, it is either the middle portion or else 
the free distal end which is split into fibrils of varying 
thickness; in the true splitting of the flagellum, leading to 
division, the basal part divides first of all into tAVO daughter- 
flagella of equal thickness, the proximal, attached ends first 
separating. In the figures mentioned, the splitting has not 
