62 Professor J. J. Thomson [March 10, 



tion. There is, however, considerable indirect evidence to support 

 the view that the atoms in many compounds are electrified. I may 

 mention, as examples of such evidence, the power possessed by certain 

 molecules, such as those of sugar, of rotating the plane of polarisation 

 of light passing through them,. This power, which is associated with 

 the presence of the asymmetric carl)on atom with four dissimilar atoms 

 attached to it, is readily explained by the electromagnetic theory of 

 light ; if the atoms in the molecule are charged, it is difficult to see how 

 uncharged atoms could produce sufficient rotation. 



Let us consider the difference in the chemical properties of a sub- 

 stance according as the atoms in the molecules are held together by 

 forces of the E or M type and one held together by the M type. Let 

 us take the molecule of marsh gas as an example, and suppose that 

 the molecule is in equilibrium under the E forces exerted by the 

 carbon atom on the negatively electrified hydrogen atoms and the 

 mutual repulsions between these atoms. The forces exerted by these 

 hydrogen atoms depend entirely on the charge carried by the hy- 

 drogen atom ; none of these forces would be affected if we replaced 

 any or all of the hydrogen atoms by any atom which carried the same 

 charge. Hence, without altering the architecture of the molecule, we 

 might replace any or all of the hydrogen atoms by atoms of any uni- 

 valent substance. In this case, the replacement of an atom by another 

 of the same valency would be a very simple thing. 



Suppose, however, that the atoms in the molecule were held 

 together by forces of the M type, then the forces between two atoms 

 would depend on the structure of both the atoms. If now we were 

 to replace one of the H atoms by an atom of another kind, not only 

 would the force exerted by the carbon atom on this atom be altered, 

 but the forces exerted by the atoms on the remaining three hydrogen 

 atoms Avould be radicaJly changed ; this change in the forces would 

 involve a complete change in the structure of the molecule. Thus 

 the effects of replacement are much more serious when the forces are of 

 the M type than when they are of the E type. The forces of the E 

 type are, I think, those which are most effective in binding atoms of 

 different kinds together, while the M type of forces finds its chief 

 scope in binding similar atoms together as in the molecule of an 

 element, or as in the connecting the carbon atoms in the carbon com- 

 pounds. 



Let us sum up the results we have arrived at. We have seen that 

 an atom built up of corpuscles in the way we have described possesses, 

 whether charged or uncharged, the following properties. There are 

 certain directions fixed in the atoms, along which or in directions not 

 too remote from which, electrified particles, positively electrified for 

 some kinds of atoms, negatively electrified for others, and either 

 positively or negatively electrified for still other kinds of atoms, will be 

 in stable equilibrium, if placed at suitable distances from the centre of 

 the atom. We may call those directions the valency directions, and 

 the regions within which the equihbrium is sbible the valency regions. 

 Those who arc famihar with the beautiful theory of Yan't Hoff and 



