Olive, Mitotic division of tlie nuclei of tlie Cyanopliyceae. 
37 
Glococapsa, however, furnislies two peculiarities in its coli 
division. Tlie cutting in two of tlie cell is accomplislied by 
simple constriction, instead of by a ring-formed wall; and, 
secondly, it lias an exceptional plane of division wliicli lias beeil, 
so far as tlie writer is aware, nowliere eise observed. Instead 
of tlie division of tlie cell occurring in a plane parallel to that 
of nuclear fission, as in normal cases, in Gloeocapsa , tlie plane 
of constriction is at right angles to tliat of tlie division of tlie 
nuclens (p. 30). 
11. Although tlie central body in vegetative filaments seems 
to be in a state of continuons mitotic activity it appears occa- 
sionally to make a beginning toward a resting condition, and 
to form a delicate membrane and karyolymph. It is probable, 
however, that tlie nuclei in tlie active filaments do not ordinarily 
approacli nearer to a state of rest than t-lie spireme condition 
or a stage immediately prior to it. This condition is not usu- 
ally attended by tlie secretion of karyolymph (p. 32). 
12. In spores and heterocysts, on the other hand, tlie 
nuclei enter a condition of rest, in wliicli nuclear vacuole and 
membrane are formed. In heterocysts, the protoplasmic Con¬ 
tents soon die, leaving nothing finally evident but disorganized 
chromatin granules. 
In some spores, the nuclear vacuole and membrane may 
persist: wliereas, in the case of Cylindrospermum and probably 
in other forms, tlie multitude of granules of reserve food en- 
croacli so upon the nuclear cavity tliat, in tlie mature spores, 
the membrane appears finally to be broken down and the gra- 
nules enter the nuclear space. We thus liave tlie peculiar 
phenomenon of an enormous central body, containing an abun- 
dance of cyanopliycin bodies, in tlie interstices of wliicli are 
the chromatin granules (pp. 27—-29). 
13. The blue and green coloring matt er s are held in a dif- 
fused state in a peripheral cliromatophore, which may have. in 
some species, the form of a liollow cylinder, or, in otliers, of 
a liollow sphere. In the six genera examined, no evidence 
whatever was found of the presence of minute „cyanoplastids u 
(pp. 14—16). 
14. The only kinds of granulär inclusions which were found 
were the cyanopliycin granules and tlie slime globules. Tlie 
cyanopliycin granules are evidently a form of stored food; and, 
in those algae with a cylindrical cliromatophore, they lie in the 
cytoplasm, generally closely packed along both sides of the 
cross partitions. In those species with a liollow spherical chrom- 
atophore, the cyanophycin bodies appear to be located in tlie 
cliromatophore itself, or, niore probably, in tlie cytoplasm be- 
tween the cliromatophore and tlie central body (p. 33). 
The slime globules are also, at least as a usual thing, lo¬ 
cated in the cytoplasm. They lie, however, in tlie immediate 
vicinity of the nucleus; so close are they, in fact, that they 
