484 



e {V — V') (in which e represents the charge of the electron;, and 

 which is derived from the available thermal energy of the medium. 

 Fulling the latter in approximation proportional lo T, we find ; 



e(V—V') = kT, 

 so that the following relation is found: 



V— V' 



— ™ — must be constant for all cyanizing reactions. 



f. Excitation potential. 



With the analogous structure of the "outer electron shells" of 

 the homologous elements it is probable that in the metal oxide the 

 dislocated electron as a rule and on an average will be in a cor- 

 responding abnormal path. With the available date we can, however, 

 not say whicii. When we, howerer, compare the values of the excitation 

 tension V" of the different elements, which quantity is decisive for the 

 energy-supply required to transfer an "outer" electron in the metal 

 atom from the normal into the first abnoiinal path (Row 5 of the table) 

 with the ionisation potential (low 3), the analogous course of these 

 \alues with increasing electro-positivity of the elements, is striking. 

 When we, therefore, introduce the quantity V — V" instead of V — V' 

 we should, reasoning in the same line, obtain practicable results 

 not only for these cases in whicli the abnormal path of the valency 

 electron in the compound would be identical with the first abnormal 

 path, but also when the position of the dislocated electron would 

 be identical with another abnormal path. A considerable difference 

 between V" and V' is, however, unlikely, because then the value 

 of the quantity V — V" would no longer be in accordance with the 

 considerable energy-supply required if the primary reaction is to 



take place '). 



V-V" ^^ 

 In row 6 the values are recorded of the quantity — — — .10'. 



It is seen that the difference between the alkalies and the earth 

 alkalies is smaller than in row 4. Considerations for a further cor- 

 rection wil be given elsewhere. 



(/. In conclusion we will remark that with the aid of Ruther- 

 fobd-Bohh's atomic model we have endeavoured in the above to 



') This is the more cogent as moreover at the complete addition of the "outer" 

 electron lo the oxygen rest energy is liberated. We have not considered this more 

 cicsely, because this increase of energy with regard to the oxygen rest may be 

 put equal for all the metal oxydes considerated, and can therefore not give 

 occasion to characteristic differences. [See also 'note at the correction"]. 



