CELLS IN DIVISION 



Dynamics of the spindle 



The living spindle is usually free of granules, a surprising fact that 

 seems to have attracted very little attention. This may be because the 

 spindle is derived, at least partly, and perhaps in some cases entirely, 

 from the relatively clear nuclear sap (Conklin^*^''). In many cells there 

 is certainly no great discrepancy between the size of the nucleus and 

 the size of the spindle; but in others, particularly egg cells, there is 

 often a considerable difference, and there can be no doubt that much 

 cytoplasmic material is included in the spindle. It is well known that 

 orientated structures tend to eject foreign bodies, and it seems likely 

 that this is the cause of the absence of granules. The layer of particles 

 that often envelopes the spindle probably consists of inclusions that 

 have been extruded in this way. The tendency to eject particles may 

 also be the cause of the expulsion of loose chromosomes from the 

 spindle, and of the well known centrifugal tendency of chromosomes 

 within the spindle. 



Recently Ostergren^^^ has made a valuable study of this and other 

 spindle problems, and has thrown new light especially on metaphase 

 equilibria. He finds that trivalents always lie nearer to that pole 

 with two attachments, and he interprets this as showing that chromo- 

 somal fibres exert a force that increases with distance {i.e. an elastic 

 tension) since the pull of the two fibres can only be balanced by the 

 pull of one when the two have shortened and the one lengthened. 



The idea that chromosomal fibres are elastic attachments between 

 the centromeres and centrioles is an attractive one, but it should not be 

 accepted in too literal a sense. Together with the tendency of the spindle 

 to eject foreign bodies, it may explain the stability of the metaphase 

 plate once it is formed, but it does not explain its initial formation. It 

 has been suggested by Druner^^^* and Wassermann^^^ and others that 

 the chromosomes are pushed on to the metaphase plate by an advancing 

 wave of gelation, and such an idea is not inconsistent with conditions 

 in many spindles, where orientated growth spreads steadily out from 

 either centriole. It can hardly apply, however, to those spindles where 

 orientated regions form round the separate centromeres, and only later 

 coalesce to give the spindle. In these cases it is not obvious why a meta- 

 phase plate is formed at all, and one might rather expect a spindle with 

 chromosomes distributed about at random. An observation of Fell and 

 HuGHES^^^ derived from a phase contrast film of early metaphase in 

 mouse cells is interesting in this connexion. Initially the chromosomes 

 are distributed at random in the nucleus, while the beginnings of 

 orientated growth can be seen round either centriole. The nuclear 

 membrane then breaks down, and shortly afterwards the chromosomes 

 move suddenly and more or less simultaneously, on to the spindle 



125 



