150 Mr Gregory, The Abortive Development 



While these changes have been taking place in the nucleolus 

 the reticulum has given rise to a closely coiled spireme thread, 

 which, at first evenly distributed throughout the nuclear space, 

 subsequently contracts towards one side of the nucleus, under- 

 going a longitudinal fission throughout its entire length. At this 

 stage the nucleolus is often seen to be saucer-shaped, the convex 

 side being applied to the nuclear wall in the region toward 

 which the contraction of the thread takes place (Fig. 2)*. The 

 polarity of any nucleus has no apparent relation with that of its 

 neighbours. 



The contraction of the spireme, with its consequent formation 

 of a series of loops, is followed by the segmentation of the thread 

 into chromosomes. The whole process of the formation of the 

 chromosomes is essentially identical with that described by Farmer 

 and Moore f. In certain Ferns, in which a similar process was 

 found to obtain, the large number of small chromosomes made it 

 difficult definitely to ascertain that each individual loop stood in 

 a definite relation to one chromosome J ; in the Sweet-Peas the 

 positions assumed by the chromosomes at their first separation 

 must lead to the conviction that, not only does each loop give rise 

 to a chromosome, but that the apex of the loop is the actual point 

 at which, in metaphase, the limbs of the chromosome are destined 

 to separate from one another (Figs. 5 and 6). 



The chromosomes shorten up very much prior to metaphase, 

 and each appears as two short thick rods lying parallel with one 

 another. No instances of ring-shaped chromosomes have been 

 observed. 



The form of the chromosomes at metaphase renders -their 

 structure difficult to observe, but it ma}^ be referred to the typical 



A-shaped figure obscured by the shortness and thickness of the 



arms of the V (Fig. 20). In anaphase, a section transverse to the 

 spindle shows the chromosomes arranged in a ring. They are 

 V-shaped, but very thick at the apex of the V, which is turned 

 toward the centre of the spindle (Fig. 21). 



The second (homotype) mitosis is of the usual type, and need 

 not be described in detail. 



The reconstitution of the daughter nuclei from the chromo- 

 somes can be followed with great clearness in the young pollen 

 grains. 



Wager has described the fusion of the chromosomes first of all 

 into a number of small nucleolar masses, connected together by a 



* Cf. Zimmermann, Morph. u. Physiol, des Pjl. Zellkernes, Jena, Gustav Fischer, 

 1896, p. 69. 



t Roy. Soe. Proc, 1903, Vol. 72, p. 104, and Q.J. M.S., 1904, Vol. 48, p. 489. 

 J Annals of Botany, 1904, Vol. 18, p. 447. 



