CELLS IN DIVISION 



strands became more marked in the spindles of Chaetopterus eggs that 

 were kept under a coversHp for a long time. 



Very little is known of the development of orientation in the spindle, 

 although it has been realized for a long time that focal points such as 

 the centrioles and centromeres are responsible. Classical cytology has 

 been much concerned with such problems as whether the centrioles 

 contain granules, and whether they can arise de novo; regrettably little 

 is known, however, of the mechanism by which the orientation is 

 produced. Almost the only significant fact that has emerged is that the 

 orientated region round a centromere or centriole starts by being small 

 and later grows larger by a process of growth or 'crystallization'. 



The work of Swann^^* on the variation of coefficient of birefringence 

 with distance from the centrosome, has already been mentioned. He 

 noticed that the rise in this coefficient at the edge of the centrosome was 

 not as sharp as might have been expected, and concluded tentatively 

 that the orientation is built up gradually over a region of several 

 microns. The significance of this conclusion is obscure, and as Swann 

 points out, it must be treated with some reserve. The maximum coeffi- 

 cient of birefringence in the sea-urchin spindle or aster is reached at 

 about 5 or 6[i from the centre. Beyond this, there is a rapid fall, which, 

 as was pointed out earlier, may be due to a progressive change in 

 molecular and micellar arrangement, or to a thinning out of fibrils 

 radiating from the centres. In the first case the centrosome is presumably 

 a body which causes general orientation round itself; in the second it 

 must actually generate submicroscopic fibres. 



Any individual spindle is, of course, the result of orientation produced 

 by several different centres, namely the centromeres, and in most 

 spindles, the centrioles or centrosomes as well. Because the chromosomes 

 are often small and numerous, the build-up of orientation round the 

 centromeres is seldom clearly visible. The part played by the centrioles 

 is usually much more obvious, as for instance in the sea-urchin egg. 

 Even here, however, where the asters are enormously developed, a 

 significant degree of orientation is also produced by the centromeres 

 (SwANN^^^). The simple notion that the centromere is the point of 

 attachment of the chromosome to the spindle is therefore only part of 

 the truth. The orientated region generated by the centromere does, of 

 course, fuse with the rest of the spindle and thus forms the chromosome 

 attachment. The regions of orientation produced by the centromeres 

 are at the same time collectively involved in generating the whole 

 spindle; in some instances they may even be solely responsible. 



The generation of the typical form of the spindle is interesting. Where 

 there are two centrioles, the reason for the characteristic shape is simple 

 enough. When the two regions of orientation meet, it is possible that 

 both systems will continue without regard to each other; but if there is 



123 



