STUDIES ON FERMENTATION. 359 



The blue colour of the liquid in the bottle is then very carefully 

 broug-ht to a yellow tint, by running in, drop by drop, the 

 hydrosulphite with which the burette D is filled. 



We next pour 50 c.c. of distilled water saturated with air 

 into the funnel a, and pass it into the flask ; the blue colour 

 reappears. We must then bring back the colour to exactly the 

 same tint of yellow. Let n represent the number of divisions 

 on the burette denoting the volume of hydrosulphite employed 

 for this purpose. 



We repeat this last operation immediately, taking 50 c.c. of 

 the wort, the oxygen of which we wish to determine, having 

 first introduced into the funnel a a sufficient number of drops 

 of the ammoniacal solution to neutralize the acidity of the 

 woi't. Let n represent the number of divisions of hydrosulphite 

 employed to restore the yellow tint in the case of the wort. 



We once more perform the experiment with 50 c.c. of satu- 

 rated water ; let n" be the number found.* 



The ratio which the quantity of oxygen held in solution in 

 the wort bears to the quantity of oxygen contained in the same 

 volume of water saturated with air, at the temperature t, and 



under the pressure H, will be ^; it will be sufficient in 



n + n 



2 



most cases to bear in mind this ratio. 



When we want to deduce the absolute quantity of oxygen 

 held in solution in a volume V of the wort, we have merely to 

 multiply this ratio by the quantity of oxygen contained in the 

 same volume of water saturated with air, at the temperature t 

 and under the pressure H, a very simple problem if we know 

 the coefficients of the solubility of oxygen in water at difierent 

 temperatures. These coefficients are given for ordinary tem- 

 peratures in the following table, which was compiled by Bunsen. 

 We have restricted the numbers to three places of decimals : — 



* The numbers n and n" will vary as the wort, or liquid which we have 

 to test, is perfectly neutral or otherwise. Should it be acid n ^n, should 

 it be alkaline n /_ n". This would be a very exact method of estimating 

 the acidity or alkalinity of any coloured liquid. 



