THE MITOTIC CYCLE 



is clear that down to a level of size well below 2,oooA, the spindle is 

 homogeneous, and does not consist of fibres in the ordinary sense of 

 aggregates of high density. On the other hand, it is also clear that some 

 regions of the spindle may be more birefringent than others. A study of 

 the variation of coefficient of birefringence with distance from the 

 centrosomes suggests that the whole achromatic figure may not be 

 uniformly orientated, and that distinct unbranching fibrils may radiate 

 from the centrosomes, the material in between being unorientated. In 

 view of the electron microscope evidence, however, these fibrils, if they 

 exist at all, can only be some tens or hundreds of Angstrom units 

 across, so that the birefringent strands found by Inoue must be caused 

 by the grouping of numerous fibrils into bundles. It may be signi- 

 ficant that fibrils of some hundreds of Angstrom units in diameter are 

 known to exist in muscle, flagella, sperm tails and elsewhere, and that 

 in some Protozoa the centrioles generate the spindle during mitosis and 

 the flagella during early interphase. Since, however, it is not possible 

 to be certain of the existence of submicroscopic fibrils without direct 

 evidence from the electron microscope, an alternative explanation is 

 possible, in terms of an orientation throughout the spindle with mole- 

 cular and micellar arrangement varying from point to point. 



The appearance of fibres in the spindle after fixation is easily accoun- 

 ted for on either of these two systems. The eflfect of most coagulating 

 fixatives is essentially to remove the water layer round protein mole- 

 cules, so that they can approach each other and precipitate. The 

 precipitated chains may then form fibres. If the whole spindle is 

 orientated, it is to be expected that the more highly orientated regions 

 will produce well marked fibres on fixation, while the less orientated 

 regions will produce fibres that are thinner and less clear cut. If, on the 

 other hand, the spindle consists of submicroscopic fibrils in an un- 

 orientated medium, it is still to be expected that large bundles of fibrils 

 will produce visible fibres on fixation. Where the fibrils are more widely 

 spaced, they will not aggregate so readily on fixation, and here the 

 fibres produced will, as before, be thinner and less clear cut. 



Since all elongate particles possess a tendency to align themselves 

 parallel, it is not surprising that an aggregation of spindle material, 

 somewhat comparable to that produced on fixation, is occasionally 

 found in life. Cooper^^^ j^^s reported that spindle fibres are visible 

 with ordinary light in living Pediculopsis spermatocytes, and Lewis 

 (M. R.)2^^ has found that they can be produced reversibly in the cells 

 of the chick by treatment with acid. In both these cases there must have 

 been a sufficient aggregation of material to produce a considerable rise 

 in refractive index. In the same way, Hughes and Swann^^^ found that 

 delayed metaphase spindles in the chick, examined in polarized light, 

 appeared more fibrous than usual. Inoue^^^ found that the birefringent 



122 



