74 OUTLINES OF EVOLUTIONARY BIOLOGY 



spindle more or less completely disappear though the centro- 

 some may certainly persist in some cases if not in all and the 

 newly constituted nucleus (Fig. 31, G, &) enters upon a longer or 

 shorter period of inactivity accompanied by growth. 



In the meantime the cytoplasm which constitutes the cell-body 

 has also divided into two parts in a plane which passes through 

 the middle of the nuclear spindle and at right angles to its 

 length. In animal cells this division is usually effected by a 

 constriction which starts from the outside (Fig. 31, F, G) and 

 in plant cells by the deposition of a cell-plate (Fig. 34, E, c.p.) 

 in the equator of the nuclear spindle. This cell-plate forms the 

 foundation of the double cell-wall which will separate the two 

 daughter cells ; it must not, of course, be confounded with the 

 equatorial plate formed temporarily by the chromosomes. 



Various attempts have been made to explain the dynamics of 

 this remarkable process of mitosis or karyokiriesis, the essential 

 features of which are always much the same though the details 

 vary considerably in different cases. The centrosomes, with 

 their centrospheres, asters and spindle, sometimes spoken of 

 collectively as the achromatic figure, are usually regarded as a 

 special mechanism for bringing about the equitable partition of 

 the chromatin substance between the two daughter nuclei. This 

 substance is evidently so important that no rough arfd ready 

 division will suffice. It is probable, as we shall see later on, that 

 the chromomeres of which each chromosome is composed have 

 different properties, and that it is necessary, in ordinary cell- 

 division, not only that the chromosome as a whole shall be 

 divided into two parts but that each daughter nucleus shall have 

 its share of each individual chromomere (compare Fig. 77). In 

 other words a qualitative as well as a quantitative division of 

 the chromatin material has to be effected, and this is secured by 

 the longitudinal splitting of the chromosomes. A transverse 

 division would only result in the separation of the chromomeres 

 into two groups, but the longitudinal division involves each one. 



According to some observers the fibres of the nuclear spindle 

 are actively contractile and actually pull the two halves of each 

 split chromosome asunder. Others maintain that the centro- 

 somes attract the chromosomes in somewhat the same way as the 

 poles of a horse-shoe magnet attract iron filings sprinkled 

 between them. 



Of late years the electro-magnetic explanation has been coming 



