Loss of Charge by Evaporation. 493 



We have seen that a loss or diminution of the negative 

 charge on the water in V gives a positive deflexion on the 

 electrometer. But the positive deflexion obtained in the first 

 of these two experiments, with V empty, has not been in- 

 creased in the repeated experiment with V full of water. 

 And in the experiments at a lower potential with water in V, 

 although V was perfectly insulated from the enclosing box, 

 the two remained at the same potential throughout the dura- 

 tion of the experiment. We may therefore conclude that 

 there has been no loss of charge from the water by evapora- 

 tion of sufficient amount to be detected by this method of 

 experiment. 



We have, therefore, next to determine the minimum loss 

 of charge which could have been detected. To do this we 

 require to know the capacity of the system consisting of 

 quadrants, wire, and vessel V. By comparison with a con- 

 denser consisting of two coaxial cylinders of known dimen- 

 sions this was found to be 83 electrostatic units. 



Take I division per minute as the minimum deflexion 

 which could be relied upon. A Clark cell gave a deflexion 

 of 31 divisions; that is 1 division is equivalent to a potential- 

 difference of 1*434/31 of a volt. Let q denote the corre- 

 sponding loss of charge from V per second per unit area. 

 Then, since radius of V is 4 cms., 



_ 1 1 1-434 J_ _J_ 



q ~6Q X b X 31 X 300 X& X tt(4) 2 ' 

 = 0'84 x 10 -6 electrostatic units. 



Thus a loss of charge from V of 0'86 X 10~ 6 of an electro- 

 static unit per second per square centimetre could have been 

 detected. 



2. Evaporation of Water at High Temperature. 



Arrangements were now made to experiment with water 

 heated up nearly to the boiling-point. Readings were not 

 taken when the water was actually boiling as electrical effects 

 due to bubbling might then be introduced, and also because 

 it has already been found * that the steam rising from boiling 

 water charged with electricity carries off" none of that 

 charge. 



For these new experiments a slight change in the apparatus 

 of fig. 1 had to be made. There the insulating supports of 

 the plate C and of the platform B are inside the case A ; but 



* Beattie, Phil. Mag. xlviii. 1899, p. 97. 

 Phil. Mag. S. 5. Vol. 50. No. 306. Nov. 1900. 2 M 



