452 PHYSIOLOGICAL CHEMISTRY 



some instances where the amount of acetone present is very small it 

 is necessary to allow the tube to stand 24 hours before making the ex- 

 amination for iodoform crystals. This test serves to detect acetone 

 when present in the ratio i : 100,000. 



3. Sodium Nitroprusside Test (Legal). Introduce about 5 c.c. of the urine or 

 distillate into a test-tube, add a few drops of freshly prepared aqueous solution 

 of sodium nitroprusside and render the mixture alkaline with potassium hydrox- 

 ide. (Be sure to add the nitroprusside before the solution is rendered alkaline.) 

 A ruby-red color, which may be due to creatinine, a normal urinary constituent, 

 or to acetone, or to acetone and creatinine is produced (see WeyPs test, page 

 402). Add an excess of acetic acid and if acetone is present the red color will 

 be intensified, whereas hi the absence of acetone a yellow color will result. 

 Make a control test upon normal urine to show that this is so. 



A similar red color may be produced by paracresol in urines con- 

 taining no acetone. 



Two hypotheses have been proposed to explain the color reaction 

 between acetone and nitroprusside: (i) The formation of a complex ion 

 of ferropentacyanide with the isonitroso compound of the ketone, 

 or (2) the formation of such an ion with the isonitroamine derivative 

 of the ketone. 1 



4. Iodoform Test (Lieben). Introduce 5 c.c. of the urine or distillate into a 

 test-tube, render it alkaline with potassium hydroxide and add 1-2 c.c. of iodine 

 solution drop by drop. If acetone is present a yellowish precipitate of iodoform 

 will be produced. Identify the iodoform by means of its characteristic odor and 

 its typical crystalline form (see Fig. 4, page 30). 



While fully as delicate as Gunning's test (2) this test is not as 

 accurate since, by means of the procedure involved, either alcohol or 

 aldehyde will yield a precipitate of iodoform. This test is especially 

 liable to lead to erroneous deductions when urines from the advanced 

 stages of diabetes are under examination, because of the presence of 

 alcohol formed from the sugar through fermentative processes. 2 If 

 protein is present in the urine to be tested it may prevent the acetone 

 from responding to the above reaction. It is therefore advisable to use 

 the distillate to secure most accurate results. 3 Sobel 4 has suggested a 

 quantitative method for acetone based on Lieben's test. 



5. Salicylaldehyde Reaction (Frommer). Render 10 c.c. of urine strongly 

 alkaline with potassium hydroxide, add 10-12 drops of a 10 per cent solution of 



1 Cambi: *Atti. accad. Lincei, 22, 376, 1913. 



2 Welker reports the production of a pink or red color during the application of this test 

 to the distillates from pathological urines which had been preserved with powdered thymol. 

 He found the color to be due to an iodothymol compound which had been previously pre- 

 pared synthetically by Messinger and Vortmann. 



3 Rosenbloom: Jour. Am. Med. Ass'n, 59, 445, 1912. 



4 Sobel: Schweiz. Apoth. Ztg., 52, 62, 1914. 



