45 



used (Vclcy^-). The period of quiescence is followed by periodic 

 cvolutiou of gas, and the duration of the jieriod is, jierhaps, a measure 

 of the rate of growth and number of gaseous nuclei formed in the 

 liquid. No quiescence was observed with solutions of peptone and 

 ferric hydroxide, but after an initial rapid effervescence the process 

 went on as in the case of the other solutions. An important contri- 

 bution to our knowledge of the law governing the rate of escape of 

 gases from solution was made by Carlson^ (1911), who passed an 

 indifferent gas over a well-stirred solution of oxygen in water. This 

 method of investigation (Bohr^) enables the results obtained, as 

 Meyer»3 pointed out, to be interpreted in the Ught of the Nernst*'- 

 diffusion theory. Perman** and later Steele^* showed that the rate 

 of removal of carbon-dioxide bj' a stream of air, was proportional 

 to the concentration of the gas in solution. Both the apparatus 

 employed and the method of interpreting results are weU suited for 

 the investigation of colloidal solutions. In the only investigation of 

 this nature which has been reported, Ftndlay and King" used 

 mechanical shaking and supersaturated solutions of carbon dioxide 

 (saturated at two atmospheres and reduced to one). Experimental 

 conditions so obtained are more in accordance with those encountered 

 in brewing, and it was shown that the gas effervescence is considerably 

 influenced by both the quantity and nature of the disperse phase. 



Using water as a standard of reference the degree of supersaturation 

 was proportional to the rate of evolution* and inversely proportional 

 to the coefficient k in the equation — 



Velocity = k x (degree of supersaturation). 



By plotting the values of velocity coefficient against degree of 

 supersaturation a means of representing distinctly the characteristic 

 behaviour of colloids is obtained. Hugo Muller*i observed that 

 freshly carbonated water loses its gas more readily than a solution 

 which has stood. This was not confirmed, for all observations on 

 water k is almost constant and directly proportional to degree of 

 supersaturation. 



There is a much more rapid evolution of gas initially from solutions 

 of gelatine, peptone, ferric hydroxide, and agar, than from water 

 solutions of potassium chloride, dextrin, starch and platinum sol 

 (dilute), in fact, a -05 per cent, gelatine sol effervesces as rapidly 

 as a 3 per cent, dextrin or starch solution and a httle more than half 

 as rapidly as a -7 per cent, peptone sol. Towards the end of the 

 effervescence, i.e., as the degree of supersaturation diminishes, in 

 the gelatine and agar solutions, the velocity coefficient increases 

 — the more so the greater the concentration. But there is a marked 

 falling off in the value of the coefficient for solutions of dextrin, 

 starch, peptone, ferric hydroxide and suspensions of charcoal. Such 

 decrease is attributed to the slow outward diffusion of the gas 

 dissolved in the disperse phase. With carbon the effect was observed 

 only when the suspension was kept in contact with the gas for a 



* That the logarithmic law holds true for rate of evolution from super- 

 saturated solutions is confirmed by calculations iiom the Pressure recovery 

 curves of Patten and Mains*' for carbonated water. 



