1-t Olivr, Mitotit- (livisioii uf (lu- imclci ni' 1 lic ( 'yauupliyceae. 



the present paper. Botli are cross sections of O.scülatoria prin- 

 ceps, draAvn witli canii'ra to about tlie same scale. Both are 

 stained witli iroii liaoinatoxyliii and show inucli the same f'eatim^s, 

 It will be noticed, howevor, tliat the vacuoles in Fischer\s ligure 

 are drawn perfectly round instead of angular and the chromatin 

 o-raniiles are relatively too small, facts which I have supposed 

 inio;ht be attribntable to the carolessnoss of the lithogra])h('r 

 to Avhoni the making of the drawings was intrusted. It is triily 

 remarkablc, however, that Fischer did not at least attempt to 

 give an explanation of the minute black graniües shown in this 

 tigm-e as well as in ügure 37, other than simply to call them 

 ..Granulationen". Even a casual glance will show that they ai-e 

 not the same as the relatively much larger granules of his 

 tigure 41b, e. g., or, indeed, of any of his figures of granulär 

 inclusions. How the fact that the minute granules shown in 

 tigure 36 stain exactly as do the chromatin granules and that 

 tliey have in addition other characteristics whioh belong to the 

 chromatin of nuclei, could have escaped so noted an observer 

 as Fischer, is to the writer inexplicable. 



Wager, in his recent preliminary paper, shows apparently 

 only one iigure whicli is drawn from a section, and, in tlie 

 opinion of the writer, this cross section was probably strongly 

 overstained. At any rate, in my own iron haematoxylin prepa- 

 rations, I can find many similar appear, ances, resulting from a 

 failure to wash out sufficiently. In fact I have seen, in my 

 own preparations, satisfactory explanations for such misleading 

 appearances as are shown by Wager in his figure 1, which 

 lead liim to the conclusion that the division of the nucleus is 

 direct. And, moreover, in thick, deeply stained sections, one 

 may find similar figures to those given by Kohl — figures 

 10 — 12, and 16, of Plate e; 14 and 15 of Plate f ; and the most 

 of the figures of Plates i and k, — to prove the opposite con- 

 clusion that the division of the nucleus is indirect. It is easy 

 to find, in overstained or badly fixed mounts, such long streaks 

 of blended chromatic and iibrous achromatic elements (see 

 figs. 2, 3) as are shown by the figures of both Wager and 

 Kohl, and wdiich are interpreted by the latter as the chromo- 

 somes of a mitotic figure, and by Wager as the chromatin 

 granules of a simple amitotic division. Overstained or poorl}^ 

 fixed preparations and attempts to fathom from without instead 

 of examining from thin sections the internal structure of a cell 

 which contains at least three different kinds of granulär in- 

 clusions and a protoj^lasmic structure showing considerable 

 amount of differentiation, must be held in the main responsible 

 for the extreme confusion and conflicting results with which we 

 are confronted. 



The coloriiig- matters. 



As was first pointed out by Schmitz, in 1879, a close 

 examination of one of the blue green algae reveals the fact 

 that, even in the living condition, wo may distinguish two 



