NO. II STRUCTURE OF THE ATOM PARSON 37 



more frequently be able to take up configurations such that they can 

 attract one another. It must be remembered, of course, that, from 

 the very nature of magnetic forces, no groups of magnetons could 

 affect one another appreciably at distances that are much greater 

 than the distances between the magnetons within the groups, for 

 changes such as could decrease the mutual energy of the groups must, 

 except for slight changes, increase their internal energy still more, 

 because the distances there are smaller. However, intra-molecular 

 distances would not usually be too great, for we have assumed the 

 radius of the magneton to be of the same order of magnitude as 

 that of the atom. 



We have now reached the following generalizations about the 

 mutual action of groups of magnetons : 



i. Groups of eight can attract one another and irregular groups 

 can attract one another much more than irregular groups can attract 

 groups of eight. 



2. A group of eight will tend to induce the formation of other such 

 groups in its vicinity ; and conversely, an irregular part of the mole- 

 cule will tend to weaken any groups of eight that are near to it. 

 Groups of eight will also mutually reinforce one another. 



These principles are of great promise in connection with the 

 influence of " negative " groups in the molecules of Carbon com- 

 pounds, for the negative action of an atom has been identified, in the 

 preceding pages, with its tendency to form a group of eight. Another 

 application is to the properties of unsaturated molecules (§§12, 13), 

 for these will naturally show the disturbing influence of free mag- 

 netons on groups of eight, if the present conclusions are correct. 



There are also residual electrostatic forces to be considered. It 

 has long been recognized that the bond in a molecule like HC1 is 

 electrostatic, and owes its existence to the extraction of an electron 

 from the H atom by the CI atom, whatever may be the cause of that 

 extraction. The electrical polarity which presumably is thus set up 

 in the molecule has been used to explain many phenomena by Sir 

 J. J. Thomson in a recent paper on " The Forces between Atoms and 

 Chemical Affinity " (Phil. Mag., May, 1914) ; also a discussion of this 

 effect from a more chemical standpoint is given by G. N. Lewis in a 

 paper on " Valence and Tautomerism " ( Journ. Amer. Chem. Soc, 

 1448-1455, 1913), and by others. Now the explanations of the mag- 

 nitude of the dielectric constant, extent of molecular association, and 

 other things, by means of this conception are not affected by the 

 assumptions of the present theory (except in so far as they may in 

 some cases be made more definite) ; but a part of the phenomena 



