158 Prof. E. K. Eideal on the Critical 



in the rallies o£ v u . v and v z >. together with the uncertainties 

 associated- with their experimental determination do not 

 altogether exclude the hypothesis. Likewise, if this point 

 of view were adopted, the relationship mv\. v ='M.v\ r , due 

 to Haber (Ver. d. Deutsch. Phys. Gesell. xiii. p. 1117, 1911), 

 in which m and M are the electronic and atomic mass 

 respectively, would have to be rejected ; the various spectral 

 formulae of Deslandres, Rvdberg, and Kayser indicate that 

 there is a simple relationship between the activating fre- 

 quencies, but not so simple as the one suggested by Baly. 



In addition to a knowledge of the various possible energy 

 increments and to the average potential energy which a 

 molecular species has to acquire before becoming reactive 

 for the particular chemical or physical reaction con- 

 templated, it is necessary to know the various fractions 

 of the gram-molecule which are already partially activated 

 to these different extents. Some at ordinary temperatures 

 are evidently already in possession of the full quota of 

 energy, since reactions like chemical combination and 

 evaporation are normally proceeding, if indeed but slowly ; 

 others possess less potential energy, and an inspection of 

 the absorption spectra makes it extremely plausible that 

 some molecules already exist in all the possible stages of 

 activation. These various molecular types differing only 

 in their potential energy are evidently in equilibrium with 

 one another, and the nature of this equilibrium must be 

 established before the calculations of the mean critical 

 energy movement as a sum of the various critical energy 

 increments can be undertaken. 



The distribution of the molecules anions these various 

 types may obey the chemical law of mass reaction or may 

 follow the kinetic distribution according to Maxwell's law. 

 It would appear more probable, however, that the dis- 

 tribution law of the molecules amongst the various types 

 is governed by the nature of the circumambient radiation 

 with which they are in equilibrium and follows the law 

 obtained by Planck for the distribution of energy in the 

 spectrum. 



According to this hypothesis the magnitude of the 

 various ordinates in the E, X curve would be proportional 

 to nhv, where n is the number of molecules each possessing 

 the amount of energy liv. For one molecular species there 

 would be a series of such ordinates spaced according to the 

 number of spectral lines: for "black" molecules, i.e. 

 those capable of all possible reactive frequencies, the 

 curve would naturally be complete, since the ordinates 



