DISPOSITION OF CHROMOSOMES IN MITOSIS. 1 99 



of small magnetized needles which were floated by being passed 

 through small discs of cork and placed on the surface of water 

 in such a manner that all the magnets were vertical in position 

 and similarly oriented, /. e., with all the north (or south) poles 

 uppermost. Such floating needles repel one another with a force 

 inversely proportional to the square of their distance apart. If 

 then over such a group floating with (for example) north poles 

 uppermost a large bar-magnet is suspended in a vertical position 

 with its south pole downward, all the needles are, by the hori- 

 zontal resultant of the attractive force of the large magnet, drawn 

 in toward a central position immediately below the latter ; here 

 they adopt arrangements of perfectly definite and regular con- 

 figuration, whose exact form varies with the number of needles 

 but shows great constancy for any given number. 1 The condi- 

 tion of stability of such systems is that the mutual repulsion which 

 the small magnets exercise upon one another shall be exactly 

 balanced by the centripetal attractive force due to the large 

 magnet. 



In their manner of grouping such floating magnets exhibit a 

 close resemblance to the chromosomes of many equatorial 

 plates, for example, those of the sea-urchin egg (seen in face). 

 So far as I am aware, however, no attempt has hitherto been 

 made to explain the grouping of chromosomes as due to condi- 

 tions of the above kind. The following experiments have been 

 designed with a view to testing this hypothesis. 



To simulate the nuclear chromatic filament, the following pro- 

 cedure is employed : Small, similarly oriented, magnetized needles 

 are strung at short and regular intervals (say 6 millimetres apart) 

 along a delicate silk thread, preferably a single silk filament ; 

 each needle is passed through a small cubical piece of cork ; the 

 entire filament can then be floated on the surface of water with 

 the needles vertical in position. When left undisturbed such a 

 filament tends to be pulled out into a straight line by the mutual 

 repulsion of its units ; this form would be taken if the filament 

 were ideally flexible and the supporting fluid devoid of viscosity, 

 since then the average distance of the needles apart would be as 



1 Figures showing these arrangements will be found in any good text-book of 

 physics, for example Ganot's. 



