14 Light Absorption and Fluorescence. 



establishment o£ those molecular frequencies which are 

 multiples of the infra-red fundamental. Both of these 

 are directly connected with problems of chemical reactivity, 

 and are discussed in the following paper. One point only 

 arising from the foregoing discussion of energy quanta 

 may be mentioned here. It has been shown that in the 

 water molecule the oxygen atom exhibits two characteristic 

 frequencies and the hydrogen atom one. whilst in sulphur 

 dioxide the oxygen atoms exhibit one and the sulphur atom 

 two characteristic frequencies. It is difficult to avoid the 

 conclusion that the characteristic atomic frequency is 

 the basis of the valency of that atom. Thus a univalent 

 atom may be one for which there is only one possible 

 shift of its electrons, with a bivalent atom there may be 

 two possible shifts, and so on. From this it would also 

 follow that the numerical size of the elementary quantum 

 associated with the atoms of an element determines the 

 position of that element in the series of electropositivity. 

 Obviously, the larger the elementary quantum associated 

 with an atom, the greater will be the energy given out 

 when that atom enters into combination. Further, when 

 a multivalent atom enters into successive combination with 

 atoms of a given univalent element, its largest elementary 

 quantum will be concerned when it combines with the first 

 atom. This will be followed by the next largest, and so on. 

 This will mean that the "strength" of its different valencies 

 will be different, and the individual bonds with the various 

 atoms of the univalent element will require different amounts 

 of energy to resolve them. This is, of course, well known. 

 In spite of the interesting nature of these deductions, 

 it is perhaps premature to write of them until the under- 

 lying phenomenon of chemical affinity has been dealt 

 with. 



In venturing to put forward the conclusions given in this 

 paper, I gain some confidence from the fact that they lead at 

 any rate to exact and rigid relationships as regards the 

 energy changes suffered by a molecule. These would seem 

 to be free from that vagueness which is so often associated 

 with attempts to apply the energy quantum theory. 



