74 CELL-DIVISION 



cations and limitations, which show that the matter is by no means 

 so simple as it first appeared. The most important of these modifi- 

 cations are due to Hermann ('91) and Driiner ('95), who have relied 

 mainly on the study of mitosis in various cells of the salamander, well 

 known as extremely favourable objects for study. These observers 

 have demonstrated that in this case the spindle-fibres are of two 

 kinds which, apparently, differ both in origin and in mode of action. 

 Hermann showed that the primary amphiaster is formed outside the 

 nucleus, without connection with the chromosomes, and that the 

 original spindle persists as a "central spindle" (Figs. 21, 22), which 

 he regards as composed of non-contractile fibres, and merely forming 

 a support on which the movements of the chromosomes take place. 

 The contractile elements are formed by certain of the astral rays 

 which grow into the nucleus, and become attached to the chromo- 

 somes, as Boveri described. By the contraction of these latter fibres 

 the chromosomes are now dragged towards the spindle, and around 

 its equator they are finally grouped to form the equatorial plate. The 

 fully formed spindle consists, therefore, of two elements ; namely, (a) 

 the original " central spindle," and (b) a surrounding mantle of con- 

 tractile " mantle-fibres " attached to the chromosomes, and originally 

 derived from astral rays. In the anaphase, as Hermann believes, the 

 daughter-chromosomes are dragged apart solely by the contractile mantle- 

 fibres, the central spindle-fibres being non-contractile and serving as a 

 support or snbstratnm along which the chromosomes move. As the 

 chromosomes diverge, the central spindle comes into view as the in- 

 terzonal fibres (Fig. 22, G, H). Strasburger ('95) is now inclined to 

 accept a similar view of mitosis in the cells of plants. 



Driiner ('95) in his beautiful studies on the mechanism of mitosis 

 has advanced a step beyond Hermann, maintaining that the pro- 

 gressive divergence of the spindle-poles is caused by an active 

 growth or elongation of the central spindle which goes on throughout 

 the whole period from the earliest prophases until the close of the 

 anaphases. This view is supported by the fact that the central 

 spindle-fibres are always contorted during the metaphases, as if 

 pushing against a resistance ; and, as Richard Hertwig points out 

 ('95), it harmonizes with the facts observed in the mitoses of in- 

 fusorian nuclei. The same view is adopted by Braus and by 

 Reinke. Flemming ('95) is still inclined, however, to the view that 

 the divergence of the centres may be in part caused by the trac- 

 tion of the antipodal fibres, as maintained by Van Beneden and 

 Boveri. 



Heidenhain, finally, while accepting the contractility-hypothesis, 

 ascribes only a subordinate role to an active physiological contrac- 

 tility of the fibres. The main factor in mitosis is ascribed to elastic 



