402 Scientific Proceedings, Royal Dublin Society. 



Hence the area of water air-surface was 56-34 sq. cms. 



When a gas is dissolving in a liquid, we may assume that the rate of 

 passage of the gas into the liquid is proportional to the partial pressure of 

 the gas and the area of liquid exposed. Hence : — 



Eate of passage of gas into liquid = S.A.p., where p = partial pressure of 

 gas, A = area of surface, and S = rate of solution for unit area. 



As solution goes on, the gas in the upper layers of the water escapes into 

 the air, and the rate of its escape is proportional to the amount of gas in 

 solution ; hence if to = weight of gas per cc. in upper layer, then the rate of 

 escape of gas from liquid = f.ic.A. This gives us as the net rate of solution — 

 S.A.p. -f.ic.A.; and when equilibrium is reached, i.e., at saturation, S.A.p. = 

 f.iv.A. or Sp = fie. 



The value of " tv " is generally unknown, since the gas diffuses rapidly 

 from the surface layer of the liquid and the exact gas-content is uncertain. 

 If we keep the liquid mixed, we render A the area uncertain in general. But 

 if a method of mixing the liquid which would leave A still determinate is 

 possible, then we can calculate the rate of solution for a given area, assuming 

 that the gas remains at constant density. These conditions are complied with 

 in the case of a cylindrical bubble moving up a narrow tube. 



If V = volume of liquid, and p = density of the gas (assumed constant), 

 the rate of solution is :— 



dVw 



— = SAp - fie A 



= SAkp -fwA 

 dw SA A 



•'• dl = -v /;p - fw v' 



SA A 



which = a - bw (1) when a = — kp and b = / p-, 



--•(-9 



w - - = Ce' il w = o when t = o hence c = - -■ 

 b o 



.'. w = 7 (1 - e' ht ) when t = oo , w = 7 the saturation value. 



Equation (1) shows that plotting the values of — against w should give a 



straight line, and when the actual observations are plotted in this way, a 

 straight-line graph results (fig. 7). 



