380 Acton . — Studies on Nuclear Division in Desmids. /. 
larger Desmids, and all discussion will be deferred until these results 
can be summarized. 
Technique. 
Material was at first fixed in weak Flemming’s mixture diluted to 
half strength, as Lutman found this the most satisfactory fixing reagent for 
Closterium . It did not give entirely satisfactory results in the case of 
Hyalotheca . There was not much shrinkage of cell contents, but staining 
results were not good after using this reagent, possibly owing to the thick 
gelatinous sheath which is present in Hyalotheca. Stronger fixing reagents 
were tried, and of these Bouin’s picroformol solution was found to give the 
best results. This solution therefore has been tried for all the later work. 
Heidenhain’s iron-alum-haematoxylin was used for staining. In order to 
prevent shrinkage it was found necessary to exercise special care in trans- 
ferring the stained material from xylol to Canada balsam. The material was 
transferred from xylol to a solution of Canada balsam diluted with xylol to 
about one-fifth of its original strength. This was then allowed to evaporate 
slowly until it reached its original strength. 
Collections of Hyalotheca were made during the winter and spring and 
dividing filaments occurred in every collection. The greatest number 
of dividing filaments was present in material fixed towards midnight, 
but plenty of dividing filaments were to be found in material fixed as early 
as 4 p.m. 
Resting Cell and Nucleus. 
In the resting state the cell contains two chromatophores with the 
nucleus lying between them (PI. VIII, Fig. 2). Seen from the end each 
chromatophore is star-shaped, the centre of the star being formed by a large 
pyrenoid from which the plates of the chromatophore appear to radiate 
(Pig. 1). The starch-sheath of the pyrenoid is very large, and is made up 
of separate plates. The number of pyrenoids is not constant. Filaments 
in which every chromatophore contained two pyrenoids were of frequent 
occurrence, and filaments with chromatophores containing three and four 
pyrenoids were sometimes found. Their arrangement in a transverse row 
in the cell suggests that they are formed by longitudinal division of the 
existing pyrenoid, but no division of pyrenoid was ever observed except the 
transverse division of the pyrenoid during division of the cell ; so that it is 
impossible to state whether these extra pyrenoids arise by division of 
existing pyrenoids or de novo . 
The nucleus is spherical in shape ; though in stained preparations 
it often appears to be somewhat flattened by pressure from the pyrenoids. 
The nucleolus is comparatively large and stains deeply with chromatin 
stains. The reticulum is inconspicuous, stains faintly, and appears to have 
few, if any, chromatin granules. 
