THE AROMATIC OXY-ACIDS. 283 



carmin or purple red if the solution is treated with acetic acid in 

 excess ; on standing, this turns to a bluish violet. 



LIEBEN'S TEST. On adding a few drops of a dilute solution of 

 iodopotassic iodide to a small amount of the distillate that has been 

 rendered alkaline with sodium hydrate solution, a precipitate of iodo- 

 form develops in the presence of acetone, which may be recognized 

 by its odor on warming the mixture, as also by the form of the crys- 

 tals, which occur as hexagonal or stellar platelets. Alcohol and 

 acetic aldehyde give the same reaction. For this reason Gunning's 

 modification, although not so delicate, is sometimes preferred : 



GUNNING'S TEST. A small amount of Lugol's solution is added 

 to the distillate and a sufficient amount of ammonia to produce a 

 black precipitate of nitrogen iodide. This disappears on standing 

 and in the presence of acetone is replaced by iodoform. 



Gunning's test, like that of Legal, may also be tried directly with 

 the native urine. 



DENNIGES' TEST (as modified by Oppenheimer). The reagent is 

 prepared as follows : 20 grammes of concentrated sulphuric acid are 

 poured into 100 c.c. of distilled water, when 5 grammes of freshly 

 prepared yellow mercuric oxide (see Reynold's test) are added. The 

 mixture is allowed to stand for twenty-four hours and is then ready 

 for use. 



This reagent is added to about 3 c.c. of urine, drop by drop, until 

 the precipitate which is thus formed no longer disappears on stirring. 

 When this point is reached a few more drops are added. After two 

 to three minutes the precipitate is filtered off. The clear filtrate is 

 further treated with about 2 c.c. of the reagent and 3 to 4 c.c. of a 

 30 per cent, solution of sulphuric acid and boiled for a minute or 

 two or, still better, placed in a vessel with boiling water. In the 

 presence of an abundant amount of acetone a copious white precipi- 

 tate forms immediately ; while in the presence of traces only (less 

 than 1 : 50,000), a slight cloud develops on standing for several 

 minutes. The precipitate is almost entirely soluble in an excess of 

 hydrochloric acid. 



If albumin is present, the urine becomes turbid at once when the 

 reagent is added. In that case the test is continued as described, 

 attention being directed to the coarser precipitate which occurs later. 

 To such urines large amounts of the reagent must be added, the idea 

 being to precipitate everything that can be precipitated with the 

 reagent before heating. 



Oppenheimer claims that the test is as delicate as that of Lieben, 

 giving a well-pronounced reaction with a dilution of 1 : 20,000, and 

 being still discernible with a dilution of 1 : 60,000. As diacetic 

 acid yields acetone when treated with mineral acids, a positive result 

 is always obtained when this is present. But as diacetic acid is 

 usually found only in association with acetone, this fact does not 

 lessen the value of the test, and is an error, moreover, which is 

 common to the other tests as well. 



