338 SCIENCE PROGRESS 



(2) They may, and do, anastomose at the moment of segrega- 

 tion to form a network with rhomboidal meshes with the long 

 diagonal lying along a line of force. 



(3) They have great permanence and tenacity ; they can be 

 deflected by currents or by a stick ; if ruptured, they shorten up 

 towards the nearer pole, thickening as they do so (just like the 

 " short filaments " in some mitoses). 



(4) The explanation of the tenacity of the filaments is that 

 the force tending to separate any two adjacent particles is very 

 small, while the viscosity of the medium is very large. In the 

 same way, with the magnetic dust-paper model the friction on 

 the paper is enough to keep the threads from breaking up so 

 long as there is no shaking. 



(5) They may overlap at different levels, so as to decussate 

 or interlace. 



(6) If the mixture lies in a vessel that is rotated in a field, 

 they lag behind the medium, and form /-shaped spirals centring 

 on the poles. (By the use of a solidifying medium, such as melted 

 jelly, their position at the moment of solidification can be fixed 

 for photography.) 



Every one of these six phenomena has been observed in some 

 case of mitosis, and most of them are general. All have been 

 declared to be incompatible with the theory of the dual force 

 centred on the centrosomes ; whereas the incompatibility is only 

 with the behaviour of abstract geometric " lines of force," with 

 which we now see the plasmic chains are not identical. Thus 

 one series of objections to Gallardo's first view have now, by 

 our enlargement, fallen to the ground. 



Another difficulty has also arisen from too ready an acceptance 

 of an abstract representation of a material arrangement. When 

 the physicist discusses fields of force, he usually avoids complica- 

 tions by treating them as of unlimited extent, and as dependent 

 only on the figured centres. In an unlimited field, as we have 

 seen, all the lines of force are concave to the interpolar axis ; 

 whereas in the cell the polar radiations are often straight, or 

 even have a reversed (adosse) curve, turning their backs on the 

 spindle-chains to become concave to the ultrapolar prolongations 

 of the axis. We can shed light on this by examining theoretically 

 or practically the effect of limiting the magnetic field by a 

 boundary of highly permeable matter, such as an oval " mask" of 

 charcoal iron. In this case the astral rays, whether on paper or 



