ON VOLTA-ELECTEIC INVEKSION. 
295 
cells when heated, the author has observed that this change also 
differs at different absolute temperatures, and that the amount 
of change per degree Centigrade is different at different parts of 
the scale. The author also finds that with different kinds of 
cells the change in the strength of the current when the cell is 
heated is sometimes an increase, sometimes a diminution, and 
that the electromotive forces vary in different senses. 
4. The author was therefore led to suspect the existence of 
a Volta-electric inversion, and by employing appropriate cells he 
has discovered that this phenomenon really exists. The difficulty 
in finding such cells arose from the fact that for many combina- 
tions the temperature at which the inversion occurs is above the 
boiling-point of the fluid. The first cell with which the author 
obtained inversion was one in which a pair of plates of iron and 
of copper dipped into a mixture of the sulphate and nitrate of 
soda dissolved to near saturation in water. The current flowed 
when cold from the iron pole through the liquid to the copper 
pole ; but when the temperature was raised to near the boiling- 
point the current first ceased and then returned, but in the 
opposite direction. In this case, however, it was found that the 
iron pole had assumed the “ passive ” state in consequence of the 
action of the nitrate. Another combination not open to this 
objection was therefore sought. 
5.. A cell was made, using poles of iron and copper dipping 
into sulphuric acid of specific gravity T753. At a temperature 
of 25® C. this cell gave a current which flowed from the iron 
through the liquid to the copper, and causing a strong deflection 
of some 10® upon a common detector galvanoscope. The cell 
was then heated. At 70° C. the deflection had fallen to 2°, at 
140° C. it was less than 1°, at 150° C. the needle was deflected 
nearly 5° in the reverse direction, at 200° C. the reverse deflec- 
tion had risen to 22°, and when the liquid began to boil at about 
229° 0. the deflection had reached 45°. On cooling the liquid 
