398 TRANSACTIONS OF SECTION B. 



so that the potential energy, V, of the molecular group depends on one coordinate 

 only (say g,) ; then the above assumption as regards the ' critical ' state amounts to 

 the hypothesis that for some value c of (7,, V has a maximum value Vc. The total 

 range of values for g, are from a to 6 (say). For values from « to c the molecular 

 group is in one chemical condition ; from c to 6 it is in the other. 



For the reaction proceeding in the sense a to 6, we can show by statistical analysis 

 that 



«-aV. 



It = 



2 \/ira/"'e-«Vdg 

 a 



where k is the velocity constant, and a = -^. From this can be deduced 



ET 



'Uogjc ^ v.. - Vm + h BT 

 dT ET'» ' 



where Y,n is the mean potential energy of a molecular group between the conditions 

 a and c ; so that V^. — V„i is the ' critical ' increment of energy necessary for the 

 reaction. 



For the reaction proceeding in the sense b to a, we show that 



d log ¥ _ V, - V', „ + J ET 

 dT' ~ B.T' 



and therefore 



dlogK _ _V„, -V>„ 



dT ET- 



where K is the equilibrium constant. 



The idea that the forces involved are electromagnetic, and therefore subject to 

 modification due to a change of radiation density produced by the catalyst, leads to 

 the conclusion that a positive catalyst by reducing Vc — V,„ increases k, but decreases 



— ^M — , and vice versa for a negative catalyst, which is in agreement with whatever 



experimental data exist. It further can be shown that k varies as some power of 



e^ where 7 is the concentration of the catalyst, a result recently put forward by 

 Eosanoff as fitting the data. 



Prof. Fbancis illustrated and described an apparatus for measuring the velocity of 

 the catalysis of nitrosotriacetonamine and allied derivatives by hydroxyl ions, based 

 on determinations of the pressure of the nitrogen evolved. The method serves also 

 as a means for the determination of the concentration of hydroxyl ions. Neutral 

 salts produce curious effects on the velocity constant, the quantitative retardation in 

 the case of chlorides being nearly the same in such catalysis and in the saponification 

 of methyl acetate and the transformation of diaceton-alcohol into aceton by hydroxyl 

 ions. In small concentrations sulphates produce an acceleration instead of a retarda- 

 tion, and again there is an approximate agreement between the magnitudes of these 

 accelerations in all the above mentioned cases. 



The order in which the salts effect the velocity constant is the same as that found 

 in the lyotropic series, namely, iodides produce the greatest, and chlorides the least 

 effect. 



Dr. N. V. SiDGWicK. — It seems to me that while general theoretical conclusions 

 on the subject of catalysis are of undoubted value, our immediate need is for the 

 accumulation of more experimental data. Thus there is the question of the relation 

 between the catalytic influence and the concentration of the catalyst. Prof. Lewis 

 has assumed them to be proportional. This is no doubt sometimes true, but not 

 always. In some reactions the catalytic effect increases more rapidly than the cata- 

 lyst, in others less rapidly. In others again, as in Tubandt's experiments on the 

 inversion of menthone, the velocity increases very rapidly on addition of a trace of 

 catalyst, but seems to reach a maximum when the amount is still very small. We 

 do not yet know how these differences are related to other properties of the reacting 

 substances. The fundamental question of the form of the equation is still uncertain. 

 It is commonly written 



K = k X C" catalyst. 



