272 THE ROYAL SOCIETY OF CANADA 



3. Suspended particles (e.g. colloidal platinum) introduced into 

 the liquid rapidly lose their effectiveness in starting the bubbles. • 



4. It is almost certain that in all cases the bubbles were initiated 

 at the surface of the glass, although the location on the surface was 

 by no means constant except in tubes in which there were obviously 

 imperfections in the glass. 



5. Although carbon dioxide is nearly thirty times as soluble as 

 oxygen, the average time interval before formation of bubbles is 

 about the same for these two gases at the same temperature when 

 the supersaturation corresponds to the same equilibrium pressure. 



6. On the assumption that the bubble originates from a spherical 

 particle acting as a nucleus which the bubble just encloses, the 

 diameter of such a particle was calculated to be at most 5 X 10""^ cm. 



The Behaviour of Glass on Electrolysis 



By J. W. Rebbeck, B.A.Sc, M.A. 



Presented by Professor W. Lash Miller, F.R.S.C. 



Experiments have been carried out on the electrolysis of glass 

 at temperatures ranging from 75 to 160°C. and with differences of 

 potential varying up to 220 volts per mm. thickness of glass. These 

 experiments indicate that under certain conditions gas may be 

 developed and that by changing these the phenomenon can be re- 

 versed. Preliminary analysis of this gas indicates that it mainly 

 consists of hydrogen. Further experiments are under way with a 

 view of ascertaining the nature of the phenomenon. These experi- 

 ments were carried out under the direction of Professor J. B. Ferguson. 



The Diffusion of Hydrogen and of Helium through Silica Glass 



By Glenn A. Williams, B.S., M.A. 



Presented by Professor W. Lash Miller, F.R.S.C. 



In the following table are collected some of the results we obtained 

 with different samples of silica glass. The rates listed are the rates 

 per hour in cc. (0° and 760mm.), at which the gas in question at 



