294 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



the only kinoplasm present is the very distinct fibers of the spindles. 

 The granules appear to be aggregated into more or less spherical 

 groups (fig. 24), which give almost a flocculent appearance to the 

 cytoplasm. In addition there are scattered about many red- 

 stained granules, which are about one eighth of the diameter of 

 the chromosomes and have no special arrangement. 



In the mother-cells of Nicotiana, when the daughter nuclei 

 have become fully organized (fig. 26), there is found a spindle 

 between each pair of nuclei, that is, there are six spindles. These 

 spindles are indistinguishable from each other, in other words it 

 is not possible to tell either from the nuclei or from the spindles 

 which of the four nuclei are sister nuclei. There must be then a 

 disappearance of the homoeotypic spindles and the organization 

 of six new spindles from the cytoplasm, or the former persist and 

 there are organized four new spindles which are indistinguishable 

 from them. As far as could be determined, the latter is the pro- 

 cedure in this form (figs. 24, 25). 



Before the nuclear membranes are formed, the six spindles have 

 become organized and are indistinguishable. The individual 

 fibers are straight and parallel, the spindles not taking on the 

 inflated appearance until about the time of the formation of the 

 nuclear membranes. At first the fibers appear rather attenuate 

 at their ends, as in the homoeotypic division ; but very soon (fig. 

 26) they are quite uniform throughout, at least there is no zonal 

 differentiation. The fibers run from each nucleus to the others 

 and to the plasma membrane as well. Were it not for the arching 

 of the peripheral fibers of the spindle, it would almost seem as 

 if the fibers radiate in all directions from the nuclei, and that the 

 spindle is more or less an incidental and unavoidable consequence. 

 It seems quite likely, however, that the spindle should be recog- 

 nized as a special structure, though the other fibers, as far as the 

 evidence goes, are of similar composition and character. In 

 FIGURES 26, 27, and 28 the attempt was made to show all the fibers 

 in the section, which was fi\c microns in thickness, by changing 

 the focus progressively during the course of making the drawing. 

 A similar method was employed by Allen {\b) in his paper on Poly- 

 trichuni. It will be seen that the fibers are attached to the nuclei 

 more or less in bunches or tufts; from a single tuft there may 

 proceed fibers both to the plasma membrane and to one or more 



