948 
Proceedings of the Royal Society 
wire A, the following reactions were obtained ; the platinum wire 
A (in the part of the salt having the highest temperature), becoming 
metal positive. On reversing the action and increasing the 
temperature in the same way adjacent to the other wire B (fig. 'No-. 
1), the exact reverse result occurred in every experiment. 
When a copper wire was substituted for one of the platinum 
wires, the copper was metal positive during the quiet fusion of the 
salt at one even temperature throughout ; but on locally intensifying 
the heat as above, under the platinum wire only, that metal at once 
assumed the positive position compared with the copper, and by 
thus varying the conditions of temperature alternate interchanges 
of electro-chemical position between the metals could be produced 
at pleasure in the same fusing salt. 
The results of numerous repeated experiments are summarised in 
table A. 
General Remarks. 
Potassium Carbonate (Table A, Division 1). — This experiment 
was six times repeated, giving same results, and on equalising the 
temperature throughout the fusing salt, alternating currents con- 
tinued for a few moments until thermal equilibrium was again 
restored. 
Potassium Carbonate (Table A, Divisions 2 and 3). — This experi- 
ment was thrice repeated with same results. The positive position 
of platinum and E. M. F. was maintained steadily for ten minutes, 
or as long as the extra heat was applied under the platinum wire. 
During equalisation of the temperature throughout the fusing salt, 
alternating electro-chemical positions between the metals occurred 
for a short time. 
Potassium Nitrate (Table A, Division 3). — This experiment was 
six times repeated with the same result each time. On equalising 
the temperature throughout the fusing salt, the metals resumed their 
original positions, copper being positive. 
These experiments indicate that, by apportioning the heat applied 
to two dissimilar elements, such as platinum and copper, in the same 
fusing salt, it is possible to produce equipotential in an electrical 
sense between them, and by pursuing this still further, a complete 
abnormal interchange of electro-chemical position is obtainable. 
