588 Proceedings of the Royal Society 
We tried with the same arrangement the resistance of distilled 
water, and found it to be about 37,000 B.A. units per cubic centi- 
metre; but on carefully distilling water twice, we found the 
resistance had risen as high as 47,000 B.A. units, showing the 
great difference the least impurity made. 
2. The effects of heat on electric resistance. We experimented 
on water, sodium-chloride, and potassium-chloride — weak solutions 
of the two latter. We heated them to about 70° centigrade, and 
measured the resistance as they cooled. We found, as the tem- 
perature fell, the rate of increase of resistance increased, and the 
results, on being plotted, all described rectangular hyperbolas, as 
shown in the diagram. Since making our experiments we find 
that Professor Beetz of Munich has been making experiments on 
the same subject, using zinc electrodes and zinc sulphate, thus 
avoiding polarization almost entirely. His results and ours agree 
generally. 
3. From some phenomena we noticed we were led to try the 
effect of varying the strength of the current passing through the 
solution ; and as the result of many experiments we find that, as 
the strength of the current increases, the resistance seems to 
diminish. We note the results of two experiments on a weak 
solution of sodium-chloride and a stronger one of copper-sulphate. 
Resistance in Current. 
Resistance in Solution. 
10 B.A. Units 
950 B.A. Units. 
100 „ 
1000 
1000 „ 
1150 „ 
10,000 „ 
1390 
10 „ 
132 
100 „ 
138 
1000 
158 
10,000 
187 
We are not prepared as yet to say to what this effect is due. It 
may be due in some way to the polarization, but we cannot say for 
certain till we make further experiments. Our thanks are due to 
Professor Tait for kindly allowing us the use of laboratory and 
apparatus. 
