Nov. I, 1888] 



NATURE 



5 



It is the view of the author that in this, and in all other 

 cases, the spindle has a cytoplasmic origin, and this is in 

 agreement with his former tenets. The occurrence of an 

 almost complete spindle within the nucleus in 5". niiida, 

 before the break-down of the nuclear wall, is shown to be 

 very probably due to the entrance of cytoplasm through 

 a number of small pores ; since the wall, as seen in its 

 polar aspect, shows a sieve-like dotting from which a 

 perforation is inferred. The event in S. niiida differs 

 thus only in degree from that in S.polytccniata. Through- 

 out the whole process of division the nucleus is enclosed 

 I in a cytoplasmic mantle or pocket, which is suspended 

 Kpely in the cell-lumen by delicate protoplasmic filaments. 

 B( the two halves of the nuclear plate separate, a cavity 

 K formed— at first traversed by the uniting-filaments 

 Werbindiingsfdden) — which increases in size by a con- 

 nnuous absorption of fluid through its wall, and is 

 regarded by the author as a mechanism by means of 

 which the two young daughter-nuclei are driven apart. 

 Vox further details the reader is referred to the original, 

 and to the figures on Plate I. 



We pass on now to the typical events in the nucleus of 

 higher plants. In the "resting nucleus" (used in the 

 conventional sense only, in contradistinction to " dividing 

 nucleus ") there exists a definite, limiting layer, the 

 nuclear wall, which consists undoubtedly of cytoplasm. 

 The nuclear reticulum consists of a number of fibrils so 

 interwoven that it is difficult to say whether they have 

 fused into a genuine network, or really retain their indi- 

 viduality, and are simply in contact with one another. 

 The author is distinctly of the opinion that the latter is 

 the case, and that after a division the nuclear segments or 

 fibrils remain separate, never losing their individuality. 

 The probability of this view is greatly increased by the 

 constancy in number of these fibrils as shown especially 

 by investigation of division-stages of pollen-mother-cells 

 in Liliaceae. The number of segments is very commonly 

 sixteen, the relatively high number obtaining in developing 

 endosperm-cells being due to the fusion of the two nuclei, 

 which gave rise to the secondary nucleus of the embryo-sac. 

 Thus, in the endosperm of I.ilium Mariagon, Guignard 

 found twenty-four or more segments, though but twelve 

 or sixteen in the daughter-nuclei of the primary embryo- 

 sac nucleus. Although information on this head is 

 limited, it has been shown that where a sudden con- 

 siderable increase in the number of segments has been 

 observed there has been a previous fusion of nuclei, as 

 often occurs in the young endosperm cells of Corydalis 

 pallida. A slight increase, however, may often be due to 

 better nutrition. Absolute constancy in number of seg- 

 ments is only met with in the case of generative nuclei, 

 so far as investigation as yet shows. 



Lying between the fibrils, and adhering to them, are one 

 or two nucleoli. Bathing the fibrils and nucleoli is the 

 nuclear sap, which at this period is not stainable. The 

 fibrils consist of a non-staining substance, the nucleo- 

 hyaloplasm, in which are embedded a number of irregular, 

 strongly-staining granules, the chromatin-granules. The 

 author prefers to speak of the nucleo-hyaloplasm, with 

 -Schwarz, as Linin. The name nucleo-microsomata for 

 these chromatin-granules is here definitely abandoned^ 

 as there exists no true parallel between them and the 

 microsomata of the cell-protoplasm. In the resting 



vegetable nucleus Prof. .Strasburger finds no trace of the 

 faintly staining "bridges" described by Flemming and 

 Rabl as uniting the nuclear fibrils in the Salamander. 

 When division is about to take place a shortening of the 

 nuclear fibrils occurs, accompanied by a definite increase 

 in thickness. The chromatin-granules at the same time 

 run together into plates, separated from one another by 

 linin (nucleo-hyaloplasm). These plates of chromatin 

 grow at the expense of the linin. The fact, that this 

 takes place in Fritiilaria before the disappearance of 

 the nucleoli, precludes the possibility that the chromatin 

 grows at the expense of the nucleoli. It is probable that 

 this equal distribution of substance in the nuclear fibrils 

 insures completely similar products when the subsequent 

 longitudinal fission takes place. 



The dividing nucleus now enters on the " skein-phase," 

 and the arrangement of the fibrils may be seen with dis- 

 tinctness. At this period in many nuclei— as, for instance, 

 in the young endosperm of Fritiilaria impertalis—thQ 

 separate segments lie, for the most part, parallel, each 

 segment being loop-shaped with legs of approximately 

 equal length. The points of bending converge on one 

 side of the nucleus — its polar side ; the free ends terminate 

 towards the antipolar side. The polar side of the nucleus 

 would appear to bear a definite relation to the point of 

 convergence of the daughter-segments of the previous 

 division, and generally the line joining the polar and anti- 

 polar sides will cut the nuclear plate at right angles. It 

 is during this stage that the nucleoli disappear. Hitherto 

 they have occupied an eccentric position, lying it would 

 seem towards the polar side— this being the region least 

 occupied by nuclear fibrils. As the nucleoli disappear, the 

 nuclear sap becomes capable of staining, and the inference 

 is that this is due to the presence in the sap of the dissolved 

 nucleolar matter. The author regards it as improbable that 

 the nucleoli go to nourish the nuclear fibrils. The structures 

 to which Prof. Strasburger formerly gave the name of 

 paranucleoli, he now acknowledges to be simply nucleoli 

 late in disappearing, so that all the theoretical deductions 

 based on the appearance of those structures, by the author 

 and others, fall to the ground. 



The nuclear membrane now breaks down, the segments 

 place themselves in the equatorial plane forming the 

 nuclear plate, and the spindle makes its appearance. The 

 author at great length details the evidence in favour of 

 the cytoplasmic origin of this structure, but into this we 

 cannct enter here. The poles of the spindle are determined 

 before the solution of the nuclear wall, but they do not 

 influence the nuclear fibrils in their transpositions before 

 the breaking down of the wall. It must not, however, be 

 concluded from this that the changes within the nucleus 

 are entirely independent of the cytoplasm until the end of 

 the skein-phase. The division of nuclei within the embryo- 

 sac, which is almost simultaneous, would negative such a 

 view. The cytoplasm does not exert any directive in- 

 fluence on the fibrils until the breaking down of the wall. 

 The fibrils now depend for support on the filaments of 

 the spindle, and these are generally equal in number to 

 the segments-one to each, or, after the fission of the 

 segments, one to each segment-pair. The completion of 

 transposition and the separation of the segment halves are 

 carried out under the influence of the spindle, a certain 

 directive action of the poles being exerted : the segments 



