54 Mr. W. Sutherland on the Fundamental 



because thiophene can only be considered in connexion with 

 benzene and other ring nuclei, and allyl sulphocarbimide gives 

 practically the same result as the methyl sulphocarbimide. 



Summary of Results. 



The present theoretical analysis of thermochemical data has 

 been confined almost exclusively to those compounds whose 

 heat of formation can be separated into parts due to the 

 mutual actions of atoms which, according to the bond theory, 

 are joined by bonds, the action between atoms which, though 

 in the same molecule, are not joined by bonds being negligible 

 in a first approximation. Thomsen showed that the great 

 majority of organic compounds belong to this class, and his 

 method of analysis consisted in determining the value of the 

 various bindings in organic chemistry, a binding being a 

 pair of bonds which are joined to one another ; and in the 

 first part of this paper it was shown that in the haloid com- 

 pounds of the metals the heats of formation are so related as 

 to prove that the heat of formation is the sum of two parts, 

 one depending on the metal and the other on the metal and 

 the halogen. But there exists another class of compounds, of 

 which the ammonium salts may be taken as a type, in which 

 the mutual action of unconnected atoms becomes of the first 

 importance, for the four atoms of H in NH 4 C1 profoundly 

 affect the thermal attitude of the N towards the CI from what 

 it is towards each CI in NC1 3 ; but this class of cases, which 

 is small, has not been discussed in the present inquiry, which 

 has been, as far as possible, limited to such thermal relation- 

 ships of directly connected atoms as show themselves free 

 from large perturbations due to other atoms. 



By not allowing for such perturbations Thomsen was led 

 to assign peculiar and untenable structures to several com- 

 pounds ; but when allowance is made for them and the 

 heterodox structures are removed, the remainder of his ana- 

 lysis constitutes a pioneer advance into the region of thermo- 

 chemical law. Thomsen's method of expressing his results 

 was in accord with his discovery, for he assigned to each 

 binding a thermal value. In accordance with the results for 

 the haloid compounds of the metals I have adopted a slightly 

 different mode of expression, and have separated the heat of 

 formation of a compound US into three parts, (R), (S), and 

 /(RS) ; of which (R) depends only on R, and (S) on S, 

 while /(RS) depends on both R and S. In the haloid com- 

 pounds of the metals I was led to conclude that (CI), (Br), 

 and (1) are zero, while the corresponding parameters for the 



