-0 Olive, ]\Iitotic divisiou of the iiuclei of tlie Cyauophyceae. 



and at the other, tliey join witli a chromosome. Tlic extensioii 

 of tlie cell in longtli by osmotic l'orces, possibly combinetl witk 

 the actnal sliortening of the iibers themselves is the j)i"obable 

 cause of pulling of the divided chromosomes apart. 



It remains, then, to apply to the fibrous portion between 

 the separating chromosomes in figs. G, 8, 10, 14, 17, 37. the 

 term „connccting übers", or „central spindle^' in order to com- 

 plete our conception of the achromatic figure, in which we have, 

 especially in the short-celled species of OsciUatoria, a .,spindle" 

 Avhich is not at all spindle-shaped, but is rather in the form of 

 a more or less thick disc. This disc-shaped central body, as is 

 Seen, e. g. , in fig. 10, is iinally cut in two equatorially by the 

 ring formed wall which grows in from the oiiter wall; this pro- 

 cess will be fully discussed later. It is obvious that a single 

 centrosome would not suffice for such a peculiar, broad-poled 

 hgure. As a matter of fact, however, no structures resembling 

 centrosomes have been observed in any of the species examined. 



As we should exj^ect in such a long-celled species as Cylin- 

 drospermum, Ave find the nucleus also greatly elongated; and 

 here, furthermore, the whole karyokinetic figure has usually the 

 spindle shape seen in the higher plants, instead of the flattened- 

 disc shape of the short celled OsciUaforias (.figs. 77, 80, 84, 85, 

 89, 90). Hegler shows in Anabaena also central bodies which 

 are similar in form to those of Cylindro.spermum. 



Proof that the process of divisio» is mitotic. 



In the foregoing discussion, we have spoken of the dividing 

 central body as a „mitotic figure", and it has been pointed out 

 that this figure possesses both chromatin granules and an achro- 

 matic, fibrous substance. Simply showing that chromatin and 

 achromatin are present far from proves, however, that the pro- 

 cess of division is ' one of mitosis ; although it would seem that 

 merely the fact that the number of chromatin granules in every 

 cell is constant, should furnish sufficient proof. But if one 

 should judge solely from such ap]3earances as are shown in figs. 

 2 and 3, and from many drawings given by Kohl and others^ 

 we may, in fact, with equal right, decide with Wager that the 

 division is direct. 



The nucleus of the Cyanoplnjceae must certainly divide by 

 one of the two methods — either by mitosis or by amitosis. If 

 amitotic, as claimed by AVager and others, then the whole mass 

 of the central body must undergo a simple constriction , and 

 there should be no spindle, and no spireme arrangement of the 

 chromatin. The most essential act accompanying mitotic division, 

 on the other hand, is that the chromatin granules are eacli split 

 in two, so that the ;,daughter nuclei receive precisely equivalent 

 portions of chromatin from the mother-nucleus" (Wilson, 1900, 

 p. 70). Usually, moreover, duiing the preliminary stages of mi- 

 tosis, the chromatin granules are arranged along a more or less 

 convoluted thread, which, whether continuous or discontinuous,. 



