138 Botanisches Centralblatt. — Beiheft 2. 



It will be uoticed further from figares 4 and 5 that tlie poles are 

 often truncated, showing the individual fibril points of the vast 

 iiumber of spindle-fibres which are especially noticeable in Mag nolia. 

 In the mature spindle three arrangements of libres are present. 

 (I) those extending from one pole the other and which in the 

 centre are united into conspicuous bundles (2) those running from 

 the chromosomes to the poles and run in four bundles from each 

 chromosome. (Figs. 3, 4, 5). As they leave each individual 

 chromosome the libres are placed very near one another, then 

 separate slightly at the center and again approach each other at 

 the apex, thus assuming a spindle shape. The tliird kind are the 

 mantle-fibres or those diverging from the poles into the cytoplasm 

 toward the equatorial region. It was ascertained that the densely 

 granulär laycr of protoplasm Avhich is so conspicouous in the 

 earlier stages (Figs. 3, 4) of the spindle now (Fig. 5) fades away 

 very perceptably intimating that its connection withthe spindle is 

 a definite one. There are no centrospheres or centrosomes. In 

 stages of nuclear division such as figures 3, 4 and 5 the proto- 

 plasm stains brown, the spindle-fibres blue and the chromosomes 

 red which brings out the differentiation of all the parts very 

 clearly. 



Metakinesis and the shape of the daughter segments. 



When the chromosomes have become attached to the fibres 

 of the nuclear spindle, they move to the equator, and when ar- 

 ranged in the equatorial plate appear, with lower magnification as 

 rather thick liunps, closely crowded together (Fig. 4). A careful 

 study, however, shows ench to be composed of two segments which 

 are united so as to form a very short thick U-shaped ligure, a 

 nearly or completely closed ring, or simply two very short and 

 thick crescents. The crescent shaped segments lie with their con- 

 cave sides toward one another and form an ellipse or ring shaped 

 figure (Fig. 3). This is precisely what takes place oftentimes in 

 PodopliyWum and in Helleljorus as described by Mottier*). The 

 two latter forms are really much contracted rings. TheU-sliaped 

 chromosomes stand with their long axis perpendicular to that of 

 the spindle being attached at the point of bending. When seen 

 from the free ends therefore these short chromosomes give the 

 appearance of tetrads, whose Interpretation has led to much error 

 and confusion. Gregoire^) states that the spindle fibres may 

 be attached at any point of the length of the U-shaped chromo- 

 somes, but this does not seem to be substantiated by the later 

 investigations, nor by a single instance in MaguoUa. The ring 

 shaped chromosomes (Fig. 3) seem to lie with their long axis 

 parallel to that of the spindle when the fibres are attached 



') Beitrüge zur Kenntniss der Kerntlieilung in den Pollenmutterzellen 

 einiger Dicotylen und Monocotylen. (Jalirb. f. wi8^ Bot. Bd. XXX. 1897). 



^) Les clnesiö polliniques chez les Liliacees. (Lu Cellule. T. XVI. 2 e 

 fascicule). 



