224 CHEMICAL DISCOVERY AND INVENTION 



and hypolheses of Werner referred to in a former chapter. It has 

 not been found necessary by Barlow and Pope to suppose that 

 multivalent elements affect spheres of influence of different sizes 

 corresponding to the several extra valencies thrown out when in 

 other conditions of combination. The sphere of influence of 

 nitrogen, for example, is approximately three times the volume 

 of that of hydrogen, both in ammonia and ammonium chloride, 

 and that of sulphur is approximately twice as large as that of 



FIG. 64. 



FIG. 65. 



hydrogen, even in compounds in which the sulphur is commonly 

 assumed on good evidence to be sexvalent. 



Remembering that it is assumed that the volume of the sphere 

 of influence of an atom in a closely packed assemblage is ap- 

 proximately proportional to its valency, it is possible to replace 

 one atom by two or more atoms of a different kind without more 

 than local disturbance which does not require a remarshalling 

 of the whole, provided the total valency of the introduced atoms 

 does not exceed that of the atom replaced. Just as in ordinary 

 constitutional formulae, therefore, we may find an atom of 

 chlorine replacing an atom of hydrogen without disturbance of 



