Bonjiington JVater, near Leith, Scotland. 299 



the smallest proportion of nitric acid, however, should it even 

 amount to y^jVo^''^ P^''^' ^^^^ ^" ^'"^ diluted state be manifested 

 by a brownish tinge, which becomes much darker in propor- 

 tion as the quantity of nitric acid increases. If a bromide be 

 present instead of an iodide, the nitric acid can be ascertained 

 with equal facihty, from its imparting a greenish tinge to the 

 fluid. 50 grs. of distilled water containing half a grain of an 

 alkaline bromide to which 50 grs. of sulphuric acid have been 

 added, will indicate ^ono^^^ P^''^ of nitric acid, by developing a 

 brownish tinge, whereas a similar mixture without the nitric 

 acid will remain colourless. This is obvious, as in the first in- 

 stance, by the addition of the sulphuric acid, hydrobromic acid 

 is formed, which is colourless, but becomes again decomposed 

 by the presence of free nitric acid, whereby the fluid contracts 

 a brownish tinge. The presence of both bromides and iodides 

 does not interfere with the delicacy of this test for nitric acid, 

 the reaction is the same as if an iodide only were present. 

 However sure and delicate this test is, the presence of organic 

 matter makes it complicated, and where the organic matter 

 cannot be perfectly separated, the presence of nitric acid is 

 easily overlooked, which indeed happened in this analysis at 

 the first examination. In such cases I find the test with 

 indigo* preferable ; yet, where the quantity of nitric acid is 

 very small, it is best to submit the concentrated fluid to distil- 

 lation with sulphuric acid, as mentioned hereafter in the quanti- 

 tative analysis, by which process the nitric acid was estimated. 



It is obvious tliat a mineral water, containing a great num- 

 ber of ingredients, has to undergo a series of processes for 

 the purpose of ascertaining the individual quantity of each, 

 by which elaborate proceedings small proportions of them 

 might be lost. It is therefore of paramount necessity to have, 

 not only at the commencement of the analysis, but also during 

 its progress, a means of control, which will confirm the cor- 

 rectness of the results of the analysis, and it is chiefly with 

 this view that I have edited this paper. 



Alkaline mineral waters which have no volatile ingredients, 

 such as ammonia, are to be controlled, regarding the total 

 amount of their ingredients, by merely evaporating the water 

 and igniting the residue; but if the water contains chiefly 

 chlorides, and particularly earthy chlorides and chloride of 

 ammonium, the mere evaporation, at a temperature approach- 

 ing the boiling-point, is sufficient to drive off hydrochloric acid, 

 and even ammonia, and if the water contains organic matter 

 the control becomes still more difficult. As the Bonnington 

 water presents these difficulties, they shall be pointed out du- 

 * Rose's Ilandbuch der Jnalylischen Chcmie, Band i. p. 235. 



