Davy — New Method for the JDetermination of Nitrites. 565 



The gallic acid solution which I have employed for the determi- 

 nation of nitrites is a strong or saturated aqueous one, which, if not 

 colourless, can be easily made so by boiling it for a few minutes 

 with animal charcoal, filtering the mixture whilst still Avarm, and 

 then adding immediately to the filtrate sufficient sulphuric or hy- 

 di'ocliloric acid to render it strongly acid, which addition I hare 

 found prevents, to a great extent, the tendency of aqueous so- 

 lutions of gallic acid to become of a yellow or brownish tint on 

 keeping, which well-known property is due, as already observed, 

 to the tendency of that acid to oxidize under such circumstances ; 

 but by the addition of the acids stated, I have kept solutions of gallic 

 acid, which were even exposed to the air in open vessels, for over two 

 months without undergoing any change in colour. 



As to the standard alkaline nitrite solution, it may be readily pre- 

 pared by decomposing a hot aqueous solution of silver nitrite with 

 potassium or sodium chloride, and after the subsidence of the silver 

 chloride formed, diluting the solution to the required amount. The 

 one I employed was made, as Dr. Prankland directs, in his ""Water 

 Analysis," for the preparation of the standard alkaline nitrite solu- 

 tion to be employed in Griess's method for the determination of ni- 

 trites, which is prepared as follows : — 0.406 gram of pure silver 

 nitrite is dissolved in boiling distilled water, and pure potassium or 

 sodium chloride added, till no more silver chloride is precipitated. 

 The solution is made up to one litre, and the silver chloride being 

 allowed to settle, 100 c.c. of the clear solution is made up to one 

 litre, of which 1 c.c. is equivalent, as he says, to 0.01 m.gram of 

 nitrous anhydride (jSToOa) ; and he further adds, that this solution 

 should be kept in closely-stopped bottles, quite full. A solution at 

 least double this strength will, however, be found more convenient for 

 my test. I may also observe that I have likewise used a standard 

 solution made by taking the commercial potassium nitrite and boiling 

 it along with alcohol, which will dissolve out the potassium nitrite, 

 leaving undissolved the nitrate and other impurities ; and this alcoholic 

 solution, on evaporation and drying the residue, will furnish the nitrite 

 suitable for this purpose. 



In using this test a convenient quantity of water to employ is 

 25 c.c, which can be easily heated in a test tixbe of somewhat larger 

 capacity, along with 1 or 2 c.c. of the gallic solution, and a few drops 

 of sulphuric or hydrochloric acid, and, when the mixture has cooled, 

 transferring it to a flat-bottomed cylindrical glass, where the depth of 

 colour can be more readily determined, and compared with that yielded 

 by equal bulks of different mixtures of the standard solution with 

 distilled water. But where the amount of nitrite is very minute, it 

 will perhaps be better to use at least 50 c.c. of the water under exami- 

 nation. I should observe that the nitrates do not produce the reaction 

 described with gallic acid, and, unless they are present in large quan- 

 tity, do not affect the test ; and that it (the reaction with the nitrites) 

 appears to be uninfluenced by the presence of the different saKne and 



