of Edinburgh, Session 1883^84. 
597 
the resistance, and the intensity of the current. Three years ago I 
had the honour to describe before the Eoyal Society a little 
apparatus, hermetically sealed and purified as much as I was able to 
do from all traces of oxygen and hydrogen. There I thought that 
the current could be really reduced to zero. But since that time I 
have applied an extremely delicate galvanometer, and I have found 
that this cell also lets pass a never-ceasing current even with small 
electromotive forces. The residual current remaining under such 
conditions is indeed only the ten thousandth or hundred thousandth 
part of the current which would be produced by the same electro- 
motive force in a metallic conductor of the same galvanic resistance, 
and it goes on decreasing through weeks and weeks before it 
becomes constant, or rather oscillating about a constant mean 
value. 
Lately I have tried to shorten the time through which one has to 
wait for such observations, and to increase the intensity of the 
residual current by making the stratum of electrolytic fluid between 
two plane surfaces of platinum very thin. I have used plane plates 
of glass lying horizontally and separated by very thin little pieces 
of clean glass. The two plane plates were platinised along their 
interior surfaces, and the platinum covering of the superior plate (a 
rectangle of about 10 and 5 cm. side), which was smaller than the 
inferior, extended over a part of its upper side in order to fix on the 
upper sides of both plates two little hollow cylinders of paper con- 
taining mercury in contact with the platinum. By the mercury the 
platinum could be connected with the other parts of the circuit. 
If one brings drops of the electrolytic fluid near to the margin of 
the upper plate, they are sucked in by capillary force into the fissure 
between the plates and kept fast there. The galvanic resistance of 
the little apparatus is only a small fraction of an ohm, and can be 
neglected when compared with the other parts of the circuit, which 
contained about 600 ohms. The fluid at the edge of the upper 
plate was in contact with atmospheric air, and therefore saturated 
with atmospheric oxygen according to its density in the atmo- 
sphere. 
I was able, indeed, with this little apparatus, to get the constant 
value of the residual current after six or twelve hours, and to 
