ATOMIC DISINTEGRATION. 157 



magnets we have a triangular group of three within sur- 

 rounded by a group of seven, and this two-group system 

 lasts until there are fifteen magnets when it suddenly 

 changes to a system of three groups, one at the centre sur- 

 rounded by a pentagon of five which in turn is surrounded 

 by a group of nine. This three-group arrangement lasts 

 until the number of magnets increases to twenty-seven 

 when the four-group arrangement comes in, one at the cen- 

 tre surrounded by five surrounded by nine surrounded by 

 twelve. 



This is the way our little magnets arrange themselves. 

 If is also the way our little corpuscles would arrange them^ 

 selves in the atom if these corpuscles were at rest. It is 

 offered here as a clever experimental verification of mathe- 

 matical calculation, and serves to give us confidence in the 

 results of the mathematical reasoning we are about to con- 

 sider. Our little magnets are at rest, but if they were in a 

 state of steady motion describing in their successive rings cir- 

 cular orbits about the centre of the sphere the effect of the 

 motion would simply drive them farther away from the centre 

 without in many cases destroying the character of the con- 

 figuration. The corpuscles of our atom we deem to be in 

 such motion and they must revolve in their orbits either 

 in concentric rings or in concentric shells. Professor 

 Thomson has not yet succeeded in overcoming the mathe- 

 matical difficulties of their distribution in shells but he is 

 able to show that the same kind of properties would be 

 associated with shells as with rings. On the latter basis, 

 which is. almost as good as the former, he has for a certain 

 number of them, solved the problem of their distribution. 

 The following table shows the way in which the corpuscles 

 group themselves. The numbers range downward at inter- 

 vals of five. 



