of Atmospheric Nitrogen and Oxygen by Water. 91 



where w = total quantity of gas in solution at any moment, 

 S = initial rate of solution per unit area, 

 A = area of surface^ 

 p = pressure of the gas, 

 /.— coefficient of escape of the gas from the liquid per 



unit area and volume, 

 Y~ volume of liquid. 



This can be written 



-j- = a — biv, where a = SAp and b=f^, 



that is, the values of the constants under the conditions of 

 the experiment. The values of the b constant have been 

 calculated for each experiment, and are given in the last 

 column of Tables I. and II. It will be seen that the 

 experimental values obtained approximate to a constant ; 

 the deviations apparently indicate the magnitude of the 

 experimental errors. 



The values of b for the curves shown in figs. 1 and 2 

 are : — 



For tap-water fr — '0186. 



For sea-water b = "0250. 



These values may be taken as the mean values of the b 

 constant for the conditions of area and volume obtaining in 

 the experiments. In order to compare the results of these 

 experiments with those to be described it was necessary to 

 reduce the constant to unit area and volume, and in order 

 to do this, it was assumed that in this case, as in the case 

 of the experiments described in previous communications, 

 the rate of solution varied with the area exposed and inversely 

 as the volume. 



Since /= — r-, the value of /"was calculated in each case. 

 J A 



In the case of tap-water A was =12*62 sq. cm. and 

 V = 258 c.c; hence /=-388. 



In the case of sea- water A = 12' 6 7 sq. cm. and V-= 262 c.c. : 

 hence /= -509. 



These values are an index of the rate of solution when 

 a small body of partially de-aerated water is exposed to the 

 air, in a quiescent condition, and kept at as uniform a 

 temperature as possible. 



There was reason to think that the irregularity which 

 was noticed in these experiments might have been due to 

 changing atmospheric conditions, particularly the humidity 



