Mr. F. D. Brown on Molecular Attraction. 257 



tion, there will be a low vapour-tension; and with little attrac- 

 tion there will be a low boiling-point. It follows from this 

 that the attraction between the molecules of hydrogen is rela- 

 tively extremely small ; that in the case of oxygen and nitrogen 

 it is also very small, though probably much larger than in the 

 former case ; the attraction mutually exerted by molecules of 

 chlorine will be more considerable ; while with bromine, iodine, 

 and other liquid and solid elements it will be greater still. We 

 must not, however, confound the attraction exerted between 

 atoms of a substance with that between the molecules ; for 

 each atom attracts separately those of the contiguous mole- 

 cule, so that the attraction between two molecules of bromine, 

 for example, will be four times as great as between two atoms; 

 and generally when the molecule of a substance contains n 

 atoms, the attraction between two molecules will be approxi- 

 mately n 2 times that between two atoms. This is of course 

 even approximately true only when the distance between the 

 two molecules is great relatively to their size ; when the two 

 molecules are close together, the several interatomic attractions 

 will be exercised over very different distances, and will there- 

 fore be very unequal in amount. Nevertheless the above 

 remark enables us to see that in some cases the apparent 

 attraction, as estimated by the boiling-point, may be very 

 misleading. In sulphur, for example, of which the molecule 

 in the solid and liquid states is probably somewhat complex, 

 we have a substance of high boiling-point, though the mutual 

 attraction of the atoms may be comparatively small. The same 

 is the case with carbon and many other substances. 



Applying now the above considerations to a few actual cases, 

 we shall see that the relative volatility of different substances 

 is generally satisfactorily explained. Let us designate by 

 (hh) the attraction at unit-distance between two atoms of 

 hydrogen, by (oo) the attraction between two atoms of oxygen, 

 and generally by (r s) the attraction at unit-distance between 

 any two atoms R and S. Then, in the case of water, the mole- 

 cular attraction will be represented by 



4A(hh)+4.B(ho) + C(oo), 



where A, B, and C are factors dependent on the distances 

 which separate the atoms. Now we have seen that (h h) and 

 (oo) probably have small values; but (ho) is not small: hence 

 the attraction between molecules of water should be far greater 

 than that between molecules of oxygen, and the boiling-point 

 much higher, a result which is in accord with fact. The 

 boiling-point of water would probably be much higher than it 

 is were it not that the attractions between H and are exerted 



