NO. II STRUCTURE OF THE ATOM PARSON 55 



not " padded " into a more or less approximately spherical shape in 

 some such way as this, they would be expected to dissociate more 

 readily than they do in actual fact. This padding need not be so 

 complete as to abolish the so-called energy of rotation. 



With regard to the assumptions that have been made in this paper 

 about the existence, distribution, and compressibility of the positive 

 sphere, it might be argued that a reference of the ordinary super- 

 ficial conceptions of volume and compressibility to a more funda- 

 mental conception that is equally gross is no explanation at all. But 

 a simplification has undoubtedly been made by the present method. 



Again, it is obviously desirable to avoid a multiplicity of funda- 

 mental concepts ; but it has not hitherto been found possible to base 

 any theory of the constitution of matter upon a single concept. For 

 example, the nearest approach that has as yet been made to such a 

 fundamental concept is the electric charge. Now, in the first place, 

 this cannot explain observed mass relations except through the idea of 

 the volume of a charge. In the second place, an electric charge, if it is 

 nothing more than that, must be dissipated throughout all space on 

 account of the repulsions between the separate portions of it. Thus 

 the various ideas of the electron, the magneton here described, the 

 uniform sphere of positive electrification of Kelvin and Thomson, 

 and the minute positive nucleus of Rutherford, all imply the existence 

 of some non-electric constraint which has essentially the same kind 

 of action upon the electric charge as the forces of cohesion have 

 upon gross matter. If then the introduction of this idea is necessary, 

 as it seems to be, it is both legitimate and useful to develop it to its 

 fullest possible extent. 



§16. Summary of Assumptions, etc. 



I have attempted in the foregoing pages to show that this magneton 

 theory can go a long way towards explaining the most diverse kinds 

 of chemical combination ; and further applications, to phenomena 

 such as those of ionization (§9), the formation of complex salts and 

 molecular compounds (§13), and the structural influences in the 

 molecules of Carbon compounds (§11), have been suggested. Before 

 these latter can be adequately dealt with, however, one other factor 

 in the behavior of magnetons must be studied : this is the electro- 

 static retaining force which opposes the extraction of a magneton 

 from its parent atom (see Kelvin's analogy, §4), and is important 

 because it determines the strength of a linkage and sometimes its 

 mode of dissociation also. 



