Olive. Mitotic divisiou of tlie nuclei of tlie ( Vaiiopliyceae. «'l 



it lies; in some cases, liowever, e. g.. Chjlmdrospermum and Aita- 

 haena, tlie poles of tlie figure are more nearly pointed tlian in 

 otliers. In nearly all instances, tlie pressure of the miütitiide of 

 cyanopliycin granules wliicli lie in tlie siirronnding cytoplasm. 

 as well as tlie presence of slime globales , wliicli are usually 

 deeply imbedded in one side of tlie central body, determine 

 largely certain ])eculiarities and irregularities of its sliape. (pp. 

 18. 10, 20, 22.) 



In Gloeocapsa ^ owing to the peciiliar metliod of Separation 

 of tlie two daugliter spiremes, tlie acliromatic iigure remains 

 tbrougliont tlie anapliases so inconspicuous (see figs. 70 and 71, 

 e. g.), tliat tlie terni spindle woiild liardly be applied in tliis 

 case. Between tlie two spirenies of fig. 71, however, tliere is 

 present an acliromatic portion wliicli corresponds ob\'iously in 

 Position to tlie central spindle, wliile tlie piiUing übers (if any, 

 indeed, be necessar; ly present), must now be located in an 

 entirely iiew position, at tlie ends of tlie cell, so as to effect 

 tlie pecnliar Separation wliicli follows. (p. 29.) 



(2) Spirenie. We have, furtlier, a spireme tliread, partic- 

 ularly evident in the cells of Gloeocapsa., in wliicli separate and 

 distinct chromatin granules are usually denionstrable. Such a 

 spireme is niost appropriately called a „segmented spireme'', 

 and the chromatin granules imbedded in it, since tlieir nuniber 

 is constant for the species, and since tliere is no furtlier indi- 

 cation of transverse fission, corres]iond to the chromosomes. 

 vShould this be true, then we have the unprecedented phenomenon 

 of a chromosome niade up of a single chromatin granule, or 

 chromomere. (pp. 22, 29.) 



In the opinion of the writer, the dividing central body in 

 figs. 8 and 10. seen in section, so tliat the long axis agrees in 

 Position with the long axis of the nucleus in Gloeocapsa (figs. 

 74, 75), corresponds to a certain exteiit to the undoubted spireme 

 in the latter figures. Thus we sliould have, in Gloeocapsa^ a 

 simple, more or less spiral spirenie, placed crosswise to the sub- 

 sequent plane of divisiou of the cell, and in Oscillatoria and 

 the other forms, a convoluted spireme placed parallel to the 

 subsequent plane of division. It is higlily probable, furthermore, 

 tliat Oscillatoria and other filamentous forms also have tlieir 

 chromatin granules arranged, like tliose of Gloeocapsa^ in the 

 form of a segmented spireme. If this be true, then the cases 

 of apparent fusion of chromatic and acliromatic elements, seen 

 particularly in preparations stained with Flemming's triple 

 stain (as in figs. 7. 10, 17, 25), are misleading, and the granules 

 slionld all appear instead as sharply defined as is sliowii in 

 fig. 8. The achromatin, however, undoubtedly varies consider- 

 ably in density dui'ing the nuclear changes. During the times 

 of greatest density, the acliromatic portion sometimes stains as 

 deeply as chromatin, so tliat an appoarance of the fusion of the 

 two into a more or less solid tliread may be given. This opinion 

 is supported by the fact tliat it can be readily demonstrated by 



