RICHARDS. — CALOMEL ELECTRODE, 11 



the conditions and result, solutions of three very different strengths were 

 used. Obviously the solutions should be allowed to remain perfectly 

 quiet at their respective temperatures for a time sufficient to allow com- 

 plete establishment of the equilibrium. This time varied from several 

 hours, in the case of the normal solution, to a few minutes, in the case of 

 the most dilute solutions. Perhaps the best idea of the rapidity of the 

 adjustment of equilibrium with dilute solutions may be obtained from the 

 following seiies of observations, which is a fair sample. A tenth normal 

 solution of baric chloride was placed in the apparatus with mercury and 

 mercurous chloride, and one arm was heated to 30° in a thermostat, while 

 the other was packed in ice. 



At the very beginning of the experiment, potential = 0.0000 volt. 

 After five minutes, " =0.0200 " 



After ten minutes, " = 0.0218 " 



After twenty minutes, • " =0.0217 " 



After a hundred minutes, " =0.0218 " 



The whole apparatus was now placed in a thermostat at 30°, and a 

 new series of observations was made. 



At the beginning of the experiment, potential = 0.0218 volt. 



After one minute, " = 0.0050 " 



After two minutes, " = 0.0020 " 



After three minutes, " = 0.0008 " 



r After four minutes, " = 0.000 1 " 



After five minutes, " =0.0000 '" 



After ten minutes, " =0.0000 " 



After sixty minutes, " = 0.0000 " 



Centinormal solutions behaved in the same praiseworthy fashion, but 

 normal solutions were much moi-e unsteady, and required a very much 

 longer time. 



As was to have been expected, decinormal and centinormal solutions 

 gave a constant potential even. when agitated, for no essential amount 

 of mercuric salt is formed in these cases. 



In order to get rid of the thermo-electric error of having the two 

 junctions of platinum and mercury at different temperatures, an apparatus 

 was made which allowed them to be kept at the same temperature. The 

 mercury in each arm was connected by means of a capillary tube filled 

 with mercury to a suitable cup in the thermostat at 30°, where the two 

 wires from the electrometer were each connected. The tube for the 



