Dissociation and Electrical Conductivity. A451 
Kaperiments with Zinc Bromide. 
As far as our experiments have gone at present, the results 
with zinc bromide are similar to those obtained with the 
iodide. The current, however, alters much more rapidly 
with the time. 
In the preceding experiments, the zine chloride was not 
specially dried. When heated, it is known that HCl is formed. 
In order to see whether this would account for the leak from 
the chloride, the apparatus was gradually heated and dry 
HCl gas admitted. An appreciable current was noticed at a 
temperature of 200°, and this did not depend on the direction 
of the field. When air was drawn over the chloride and 
bubbled through AgNO, a precipitate was at once formed ; 
but if the nitrate was replaced by water and the bubbling 
took place for half an hour, no zine could be found in the 
solution. When the bubbling took place through starch and 
KI solution, no coloration was noticed, so that no free chlorine 
was present. Next the air was pulled through an insulated 
test-tube packed with cotton-wool and coated on the outside 
with tinfoil. This was connected to the electrometer 
quadrants. When the foil was charged negatively, there was 
a further charging up when the air was drawn through; when 
charged positively, a greatly increased leak. There would 
therefore appear to be an excess of negative electricity drawn 
off from ZnCl, at this temperature (360°). . 
Some experiments were next carried out with the object of 
discovering the nature of the ions concerned in the conduction. 
For this end their velocity was determined at a temperature 
of 360°. Fig. 9. 
A ay € 
A method previously used by McClelland * to measure the 
velocities of ions from arcs was used. The apparatus is shown 
in fig. 9. The substance is heated at R, and air is drawn 
* Proc. Camb. Phil. Soc. x. p. 241 (1899). 
212 
