46 THE ELEMENTS OF STRUCTURE. 



2 (d] Division and separation of the loops. Each of the loops 

 which make up the star divides longitudinally into two, and 

 each half separates from its neighbour. They lie at first 

 near the equator of the cell, but they are apparently drawn, 

 or driven, to the opposite poles (Fig. 5, 2-3). 



U (e) Diastroid. The single star thus forms two daughter stars, 

 which separate further and further from one another towards 

 the opposite poles of the cell, remaining connected, how- 

 ever, by delicate threads (Fig. 5, 3-5). 



* (/) Each daughter star is reconstituted into a coil or network for 

 each daughter cell, for the cell substance has been constricted 

 meanwhile at right angles to the transverse axis of the 

 spindle. The halves separate in the case of Prot6zoa, but 

 in most other cases, e.g., growing embryos, they remain 

 adjacent, with a slight wall between them (Fig. 5, 6). 



/ (g) Each daughter nucleus then passes into the normal resting 

 phase. The spindle and usually the centrosomes also dis- 

 appear. 



Flemming gives the following summary of karyokinesis : 



MOTHER NUCLEUS DAUGHTER NUCLEUS 



(progressive changes). (regressive changes). 



a Resting stage. Resting stage, g /(\ 



6 Coil. Coil. / 



Y c Astroid. Diastroid. e 



7>- d Division of Astroid and its loops J>- 



(metakinesis). 



(3) We are far from being able to give even an approximate account 

 of the " mechanism " of cell division. Rapidly progressive research has 

 disclosed many mysteries, but it does not explain them. The nucleus 

 is resolved into a chromatin framework and an achromatin matrix, but 

 we know the nature of neither. The longitudinal division of each loop 

 shows how thorough is the partition of substance and implied qualities. 

 The " central corpuscles," recently discovered, act like centres of force, 

 and the indescribably fine threads, which pass from around these to the 

 chromatin loops, have been credited with motive powers. Siriiilarly the 

 threads of the nuclear spindle are believed by some to draw or drive the 

 chromosomes, But we do not know. The whole process is vital, and 

 therefore inexplicable in terms of matter and motion, so long at least as 

 we do not know the secret of protoplasm. 



(4) On the other hand, Leuckart and Spencer have given 

 a general rationale of cell division. Why do not cells grow 

 much larger, why do they almost always divide at a definite 

 limit of growth ? Their answer is as follows : Suppose a 

 young cell has doubled its original mass, that means that 

 there is twice as much living matter to be kept alive. But 

 the living matter is fed, aerated, purified through its surface, 

 which, in growing spherical cells for instance, only increases 



