338 Dr. S. W. J. Smith on the 



stages of this reaction proceed with great rapidity. In the 



presence of a sufficient quantity o£ oxygen, /jl may thus 



almost at once reach a greater value than that required to 

 prevent the evolution or hydrogen*. 



It is therefore essential to experiment with a mercury 

 surface as far as possible oxygen-free, or to devise a means 

 of removing from near any mercury surface as much as is 

 desired of the mercury salt in solution. 



§ 8. Possible methods of eliminating this effect. — Imagine 

 two masses of mercury A and B immersed in the same acid. 

 Let the surface of A be one not originally oxygen-free and 

 surrounded in consequence by solution containing dissolved 

 mercury salt. Suppose that the state of the solution around 

 A has become practically steady without appreciable diffusion 

 of the mercury salt into the region round B. Let the surface 

 of B be one originally free from oxygen, and suppose that 

 no interaction has yet taken place between it and the acid. 

 Left to itself this mercury might interact with the acid, some 

 of it displacing some of the hydrogen ; but suppose that, 

 instead, it is brought into contact with the mass A at one 

 point or more without considerable change in the extent of 

 -either surface in contact with the acid. 



Well-known electrochemical phenomena leave no doubt as 

 to what will happen. Mercury will immediately begin to 

 deposit on A and to enter the solution round B. Mercury- 

 salt will in fact disappear and appear in equal quantities 

 round A and B respectively until the concentration of the 

 salt in solution round both is the same. This change will be 

 effected by a displacement of anionic ' chains ' in the solution 

 from A towards B and by some analogous process (shift of 

 electrons) in the mercury from B to A. The time taken by 

 the process to complete itself will depend upon the length of 

 the ionic chains. If these are short this time will be very 

 small compared with the time taken by A to reach the steady 

 state acquired before the contact. 



From the point of view already described this process 

 occurs in a way analogous to the expansion of a gas when 

 the external constraints permit. Here the constraints vir- 

 tually permit the expansion of the mercury salt from the 

 space round A into the space round B until the concentra- 

 tion is the same in both. The electrical phenomena are 

 incidents, not causes, of the flow of matter which takes 

 place. 



* The effect of the presence of oxygen can also be presented in a 

 thermochemical form, omitted here for the sake of space. 



