22 
Olive, Mitotic division of tlie nucleiof tlie Cyanopliyceae. 
sharply throughout tlieir entire length. In eitlier of tlie two 
Standard stains, one can, however, see them, usually very dimly 
defined, extending only a short distance from tlie central body. 
A general cytoplasmic stain sometimes gives better results Avith 
these structures; and, in fact, in tlie experience of the writer, 
was actually necessary in showing tlie fibrils in their entire 
length. 
As has been pointed out above, the dense, fibrous, achromatic 
portion of the dividing central body between the two groups of 
separating chromosomes (as, for example, in fig. 8), cannot well 
be interpreted otherwise than as the „central spindle“; and the 
fibrils that leacl from the chromosomes to the walls appear, at 
least, to function as mantle fibers. There remains, therefore, in 
presenting proofs of mitotic division, to discuss the more important 
phenomena accompanying mitosis, viz„ the fission of the chromatin 
granules and their arrangement in a spireme. 
A detailed account of the mitotic division of the nucleus 
in Gloeocapsa polydermatica will be reserved tili later, since the 
process in this species involves certain peculiarities which merit 
a special discussion. 
Hegler (01) says that in Anabaena , during the division of 
the nucleus, the m'inute chromatin granules fuse with one another 
to form a „grösseren Verbänden, deren Chromosomennatur an 
günstigem Untersuchungsmaterial nach Fixieren mit schwefliger 
Säure und Färbung mittels der angeführten Methoden durch ihr 
weiteres Verhalten beim Teilungsprozess festgestellt werden 
könnte" (p. 352), I have never seen any such fusion of chro- 
inatin granules to form chromosomes. In fact, if normal, such a 
process as the union of chromatin granules to form chromosomes 
should take place early in tlie formation of the spireme thread. 
I am certain that a fusion of the chromatin granules does not 
occur in the spireme thread of Gloeocapsa, unfortunately the only 
instance in which I can speak positively on this point. 
In the cells of Oscillatoria, we can frequently see nuclei 
which appear to be in a spireme stage (fig. 7, the loAvermost cell; 
fig, 10, the two middle cells; fig. 14, tlie four cells at the right; 
fig. 17, the lowermost cell). But particularly in cross sections, 
do we find appearances Avliich suggest at once a thickened 
spireme thread, in which the achromatic and chromatic sub- 
stances seeni to be blended (figs. 25, 42). Moreover, such nuclei 
as are shown in the tivo middle cells of figs. 8 and 10 probably 
represent in section a similar condition to the spireme stage of 
the nucleus of Gloeocapsa seen in figs. 68, 70, 74, and 75. It 
should be kept in mind, lioiinver, that, in Gloeocapsa, tlie spireme 
is in the form of a simple, more or less spiral thread; Avhereas, 
that which Ave see in section in the case of Oscillatoria, has its 
convolutions disposed in a disc-shaped figure. In both instances, 
we may see the beginnings of the longitudinal fission of tlie 
spireme, resulting in the doubling of tlie number of chromatin 
granules. In neither case, hoAvever, is there evident a subsequent 
