18o Proceedinys of the Royal Society of Victoria. . 



tried, and with ccmplete success, and I found the wliole cell 

 now water-tight even under considerable hydrostatic 

 pressure. Glass tabes were passed through the indiarubber 

 blocks, and Liebig condensers were attached to these by 

 pieces of indiarubber tubing. The inner tubes of the 

 (U)ndensers were closed at the top b}^ corks. The condensers 

 were held vertically by clamps, and the cell was thus 

 sus])ended. It was inunersed in oil to a depth of about 

 () cm. (dotted line in figure). I had to keep it hung, as my 

 bath was copper. I tested the insulation of the oil, and 

 could get no current through a very slight thickness of it. 

 When heating" a solution, air bubbles began to form aJfc 

 about 75° C. The bending of the tube was to allow" the 

 escape of these when they became large enough, as well as 

 to allow the steam to escape more readily when the temi>era- 

 ture rose to boiling-point. The conden.sers were, of course, 

 intended to keep the solution at a constant strength. 

 Observations of the resistance above 70° were made after the 

 solution had been well boiled, so that there were no air 

 bubbles to increase the resistance of the system. 



To observe the temperature, I took a glass tube of the 

 same section and thickness as that of the cell, and corking 

 one end, I partially filled it with the same solution as that 

 with which I was working, and putting the thermometer 

 into this, I put the tube in a slanting ].)Osition in the bath. 

 Under tliese conditions, I considered tliat tlie temperature of 

 the solution in the second tube ought not to differ much 

 from the temperature of the solution in the cell. For extra 

 security, however, I always kept the temperature within a 

 degree or two for several minutes, and within i degree for 

 about half a minute before taking a reading of the resistance. 

 The salt I used was ordinary commercial copper sulphate 

 which I purified by making strong super-saturated solutions 

 in distilled water, and taking the crystals which were 

 deposited before the .solution became cold. I obtained the 

 strength of each solution by weighing the amounts of salt and 

 water in it, and checked the results by taking the density 

 with hydrostatic balance, using a glass sinker, and then 

 comparing these values with a series previously obtained and 

 l>lMtted. 



I made a very great number of obsei'vations altogether, 

 but finally have drawn my conclusions from eight sets, 

 which were the last made, and on which I spent more time 

 and pains than on the others. In the results which follow. 



