320 
ZOOLOGY: KOFOID AND SWEZY 
An inspection of our figures and still more of our preparations reveals 
a great variety of positions of the organelles of the cell. Observations' 
on living cells under high magnifications in culture slides convince us 
that this variability in position is normal and not due to artefacts. 
These conditions have led Martin and Robertson^ to state that di- 
vision in Trichomonas gallinarum may occur in any one of three planes 
and that ^' these divisions may be termed respectively longitudinal, trans- 
verse, and oblique." 
During the period of mitosis the body is somewhat rounded up, and 
is constantly in more or less violent motion. In these activities not 
only the flagella and undulating membranes share but also the axostyle. 
This bends at right angles (fig. 2), curves (fig. 6), and twists about very 
energetically almost as a flagellum lashing about in the viscid cytoplasm. 
Its activity is accompanied by a considerable mobility of the cytoplasm 
with the result that the positions of the organelles, of the nuclei, of the 
blepharoplasts with their accompanying flagella and undulating mem- 
brane, and even of the axostyle itself, are subject to incessant readjust- 
ments in their relations. They may even become somewhat widely de- 
tached from one another (figs. 4-7). These conditions which Martin 
and Robertson^ have interpreted as indicating other types of division 
than longitudinal are thus due to the protean activity of the organisms. 
When, however, we seek to define the plane of division on morphological 
grounds and accept the premitotic cell (fig. 1) as indicating the normal 
orientation of the organelles within the body, we find that, if the daugh- 
ter blepharoplasts migrate equally 90°, as they normally do in mitosis, 
the nucleus of Trichomonas would be divided in a plane which in the 
premitotic arrangement is approximately longitudinal. The chromatic 
margin of the undulating membrane, the membrane itself, and the 
axostyle all spHt lengthwise, that is, longitudinally. On morphological 
grounds the division of this flagellate is therefore solely longitudinal. The 
interpretation of any other plane rests purely on variable and temporary 
positions of organelles which become movable during the later phases 
of mitosis. 
This process of division has been interpreted, with emphasis, in Tri- 
chomonas gallinarum by these same investigators^ as purely amitotic. 
The differences between their figures and our own are not very great, and 
it is quite possible that under certain pathological conditions nuclear 
appearances simulating amitosis may occur. We are inclined to the 
view that division in this species also is mitotic and that Martin and 
Robertson's^ figures are not wholly incompatible with this interpreta- 
tion. A division in which chromatin masses (chromosomes) of definite 
number and regular forms are differentiated, spHt longitudinally, and 
