758 Sir J. J. Thomson on the 



Some theories, as for example Laplace's Theory of Capil- 

 larity and Boscovich's Theory o£ the Atom, have attempted 

 to explain many physical phenomena by supposing that the 

 molecules attract each other with forces which are exceeding 

 intense at molecular distance but diminish so rapidly as 

 the distance increases that they very soon become inap- 

 preciable. The results obtained by the older and more 

 general theories have been perhaps rather disappointing, and 

 deductions which can be put to the test of experiment would 

 seem more likely to be given by more recent theories, in 

 which the atomic structure of the molecule is brought into 

 prominence, and two atoms of the chemical elements A and 

 B are supposed to exert a force C A C B cf>(r) on each other^ 

 where C A is a parameter characteristic of the element A. 

 C B one characteristic of B, and (j>(r) is a function of r tne 

 distance between the atoms. 



If the forces between the atoms arise from the corpuscles 

 and the positive electricity, they will depend on the orienta- 

 tion of the atoms and will not always be radial ; it is only 

 when we take the mean value of these forces that they can 

 be treated as radial and as determined solely by the distance 

 between the atoms. 



Sutherland was, I believe, the first to develop a theory of 

 this kind. In a long series of papers in the ' Philosophical 

 Magazine ' he worked out the consequences of supposing 

 that the Force exerted by an atom was the same as if each 

 atom contained an electrical doublet whose moment was 

 characteristic of the atom ; this doublet was carried by the 

 atom unaltered into all the compound molecules o£ which it 

 formed a part, provided the type of compound remained the 

 same, — it might, however, alter when the type of compound 

 changed. The force exerted between a molecule AB and 

 another CD is equal to 



(C A + C B )(O c + C D )0(r), 



where C A , C B , C c , C D are the moments of the doublets in 

 the atoms A, B, C, D respectively, and <f>(r) is proportional 

 to 1/r* where r is the distance between the molecules AB, 

 CD. The values of the moments, expressed in electrostatic 

 measures, found by Sutherland for some of the elements are 

 given in the following table * : — 



Element ... H. O. N. CI. Br. I. 



E1 M,Tif C l -75X10- 13 1-7X10- 18 1-8X10- 18 4-4X10- 18 5xl0" 18 6-2xl()- 18 



Moment 



u Further Studies on Molecular Forces," Phil. Mag. vol. xxxix. p. 1. 



