30 Mr W. N. Shaw, On Electrolytes. [Nov. 25, 



I have not hitherto dealt with the difficulty raised by the 

 remarkably accurate calculations of resistance of electrolytes from 

 ionic velocities based on the assumption of atomic ions and verified 

 experimentally by Lodge in the case of hydrogen. In the first 

 place however I find it difficult to realise the conception of an ex- 

 tremely large number of small bodies (atoms) moving in opposite 

 directions through an aggregation of other small bodies (the solu- 

 tion). There is also an objection to Kohlrausch's theory on the 

 ground that it assumes all the atoms of the dissolved salt to be 

 moving. It seems to me that the conception is easier if we regard 

 the whole of the solution as divided between the dissociated atoms ; 

 the effect of the electromotive force would then be to pull the whole 

 of one set of atoms with their associated molecules in one direction 

 and the whole of the other set with their associated molecules in 

 the opposite direction, and we thus get a stress shearing the one 

 set of molecules past the other set ; the result will be a relative 

 motion of the atoms carrying their loads and the mean velocity 

 will depend on the electromotive force and the viscosity ; but 

 the motion is relative; as to the absolute velocity of each set 

 the velocities in the two directions may be regarded as equal. A 

 particular atom may at one time be a dissociated one moving to 

 meet a partner with this velocity, at another time it may belong to 

 an associated molecule and be travelling with another dissociated 

 atom in the same direction as before or in the opposite ; to deter- 

 mine the mean velocity of all the cation atoms for instance, in 

 one cross-section, we must deduct the number of backward steps 

 it takes in the unit of time as part of a molecule associated with 

 an anion atom from the number of forward ones it takes either 

 as a dissociated atom or part of a molecule associated with a 

 cation; this mean velocity will be equivalent to a transference of 

 the atom through the solution. It is this mean rate of trans- 

 ference which must be always the same for the same atom in 

 very dilute solutions, no matter with what other atom it was 

 associated as a salt, and it is this mean velocity which Lodge has 

 measured. 



I think, therefore, that it may be possible to frame a general 

 theory of electrolytic action on the basis of a hypothesis of com- 

 plex molecular aggregates, dissociated in a solution, or separated 

 by the current, into ions consisting of atoms with attached mole- 

 cules ; and such a theory might explain all the various electrolytic 

 phenomena, including migration, endosmose, and the relation 

 between viscosity and resistance. I am well aware that the brief 

 sketch contained in the foregoing paper cannot be regarded in 

 any way as a complete statement of such a theory, and that 

 further application in detail is necessary before the theory can 

 claim to be satisfactory. All that I venture to say is that there 



