DILUTE SOLUTIONS AT THE FREEZING POINT. 333 



infinite dilution should be found for the temperature of the freezing point of the 

 solution. It is simpler, however, to measure all the conductivities at the freezing 

 point of water, and to refer them to the limiting value of the equivalent conductivity 

 at that temperature. The freezing point of the strongest solution used is less than 

 a tenth of a degree below zero, and, since the temperature coefficient of conductivity 

 only changes very slowly as the concentration of the solution alters, the maximum 

 error introduced by this simplification will be less than the unavoidable errors of 

 experiment. 



SECTION 6. On the Testing of the Apparatus. 



Insulation of the Electrodes. While the cell was empty it was connected up in 

 series with a dry lottery and the galvanometer. This was repeated at intervals 

 throughout the work. Sometimes no deflection could be seen, sometimes the 

 movement was just visible. In either case, the leakage between the electrodes was 

 negligible for the purposes of the experiments. 



Adjustment of the Liquid in the Cell to a Constant Volume. The cell was taken 

 to pieces, dried by a current of hot air, and put together again. About 230 to 250 

 grammes of water were run in from the large platinum filling machine, which was 

 weighed Ijefore and after the operation. The water was then cooled to within half a 

 degree of the freezing point, and the emptying machine weighed and placed in 

 position. The exhaust apparatus was next worked at a constant pressure of about a 

 foot of water, till no more water came out of the cell. The increase in weight gave 

 the water withdrawn, and this, subtracted from the weight put in, gave the final 

 weight of water in the cell. 



In order to test the constancy of this weight (i.e., the accuracy of the levelling), a 

 weighed quantity of water was introduced by means of the small filling machine, and 

 again withdrawn into the emptying vessel. If all is well, the weight withdrawn 

 should equal that added, and the weight of water left in the cell should keep the 

 same when this process is repeated. 



The volume of the cell has to be redetermined whenever it is taken to pieces for 

 cleaning or any other purpose. Details of one such determination are given below. 

 The importance of keeping the withdrawing pressure small and constant is well 

 shown by this example. In the third experiment, by an accident, the full force of 

 the water pump was used, and a variation of nearly half a gramme in the weight 

 of water left in the cell was the result. When more water was added, and a new 

 withdrawal made with the regulator, it will be seen that the right level was again 

 obtained. 



