Olive, Mitotic division of tlie nuclei of tlie Cyanophyceae. 
21 
splits tfirougfiout its entire lengtli into two exactly equivalent 
lialves“ (Wilson, p. 70). A tliird attribute of a typical mitosis 
is tbe presence of a fibrous, acliromatic mass known as a 
spindle. 
It remains, tfierefore, to be proved, first, tliat, in tlie di¬ 
vision of tlie central body of tlie Cyanophyceae , a spindle is 
present; second, tliat tlie cfiromatin is arranged, at some time 
dnring tlie process, in tlie form of a spireme tliread; and, lastly 
and most important of all, tliat tlie cfiromatin granules are 
fialved and tliat an equal number is distribnted to eacfi 
dangfiter cell. 
It is perfiaps advisable, at tlie very ontset, to discuss briefly 
tlie staining reactions of tlie cfiromatic and acliromatic elements 
of tlie nncleus. Wfiile tlie writer realizes fnlly tliat staining 
reactions sfiould by no means constitute tlie principal argument 
in support of tlie mitotic division of tlie central body, yet tfiere 
can be 110 doubt tliat a comparison witfi tlie well known results 
alreadv obtained witfi nuclear stains will be of valne. Tlie 
t j 
writer is well aware, furtfier, tliat staining reactions are fre- 
quently misleading; and tliat tlie most credible data concerning 
complex nuclear pfienomena are furnisfied by observing 
simply tlie cfianges wfiicfi take place. Tfiose most valuable 
stains, Heidenfiain's iron fiaematoxylin and Eiemming’s 
triple stain, wfien used to cfieck and to Supplement eacfi 
otfier, in my opinion, assist as perfiaps no otfier stains can, 
in tlie Interpretation of tlie complex structures witfi wfiicfi we 
fiave to deal in tlie central body. Iron fiaematoxylin gives 
generally mucfi tlie sfiarper differentiation and is tlie easier of 
tlie two to use: but tlie objection fias been rigfitly raised tliat 
otfier tfiings tfian nuclear structures may be stained by it and 
tliat, consequently, great care must be employed in drawing 
conclusions. I fiave found, moreover, particularly in tlie case of 
tfiese algae, tliat Elemming's triple stain is an exceedingly 
difficult combination to fiandle so as to obtain tlie best results; 
but, on tlie otfier band, it furnisfies ns witfi staining reactions 
wfiicfi can scarcely be doubted, in tlie Interpretation of cfiromatic 
and acliromatic elements of tlie nncleus. 
It is sufficient to say fiere tliat tlie most satisfactory details 
were obtained witfi well differentiated iron fiaematoxylin. Tlie 
minute black cfiromosomes stood out, sfiarply defined, in a bluisfi, 
or dark, or at times alniost in visible acliromatic substance (figs. 
8, 18, 26, 27, 32, 34 etc.). Flemming's triple stain gave dark 
reddisfi, or purplisfi cfiromatic structures wfiicfi were often poorly 
differentiated, tfius giving an appearance wfiicfi would lead one 
to tlie concfiision tliat cfiromatin granules and tlie acliromatic 
substance were fused togetfier (figs. 7, 10, 13, 17, 25 etc.). Tfie 
dense, acliromatic portion of tlie central body stained witfi tlie 
triple stain dark bluisfi, or purplisfi, or even reddisfi; wfiereas 
tlie übers wfiicfi lead from tlie cfiromosomes to tlie cross walls 
do not readily take any stain sufficiently to bring tfiem out 
