556 Adkney — Dissolved Gases and Fermentative Changes. 



I ought to note, before concluding this description, that it has been ascertained 

 by experiments, which are detailed in my Paper on the earlier form of the apparatus 

 already referred to, that the bubble of gas, which is sometimes retained in the 

 laboratory vessel during measurement, does not introduce appreciable errors 

 therein. 



Extraction of the Gases in Solution. 



Although it is impossible to completely exhaust a water of the gases it holds 

 in solution in vacuo in the manner I have described, it is nevertheless possible 

 to carry the exhaustion so far that the quantities of the respective gases left 

 behind in the water may be disregarded, as experience has shown that they do 

 not appreciably affect the results. 



Both oxygen and nitrogen are readily extracted from the water, and transferred 

 to the laboratory vessel for measurement by three or four exhaustions of the flask 

 during the boiling of the water in it, but the carbon dioxide is not so easily 

 disengaged, and between a quarter and half an hour's ebullition in vacuo is 

 usually necessary to effect its removal, the length of time depending on the 

 quantity of this gas in solution, and also, apparently, upon the dissolved mineral 

 matters also, a hard water requiring more boiling than a soft one. 



Owing to the slowness with which bicarbonates decompose when boiled, I have 

 found it practically impossible to determine, separately, the "combined" and 

 " nncombined" carbon dioxide in a water. But the total carbon dioxide is readily 

 estimated by boiling the water with a suitable acid, and by the term carbon 

 dioxide I shall hereafter always include the combined carbon dioxide with that 

 which may be held in solution uncombined. 



The acid I have used for this purpose has invariably been sulphuric acid. 

 When its presence has been necessary, I have always added it to the water placed 

 in the flask in the first instance, preparatory to exhausting the air from the flask, 

 as already described. 



The presence of the acid does not interfere with the determination of the 

 oxygen and nitrogen, unless, of course, nitrites are present in the water. When 

 this is the case, the dissolved oxygen and nitrogen are determined in the absence 

 of acid, and the carbon dioxide afterwards estimated by a separate experiment in 

 the presence of acid. 



I have generally employed 250 c.cs. to 500c.cs. of a water for the analysis of 

 the dissolved gases ; such volumes of water giving suitable volumes of gases for 

 the apparatus. The operation of extracting and analysing the dissolved gases in 

 a water may generally be completed within two hours. 



