278 PHYSIOLOGICAL CHEMISTRY 



24 hours are allowed for precipitation of the proteins to become complete. The 

 solution is filtered through a dry folded filter into a measuring cylinder without 

 washing the precipitate. The volume of filtrate is noted and is taken for analysis 

 as an aliquot part of the total blood-alcohol mixture. The filtrate is then con- 

 centrated to a volume of 3-5 c.c. and used for determination of amino nitrogen by 

 the Van Slyke nitrous acid method (see Chapter IV on Proteins). The use of a 

 few drops of caprylic alcohol to prevent foaming is advisable. 



(b) Method of Constantino. 1 This is based on the formol titration procedure. 

 One hundred c.c. of blood or serum is mixed with a measured (500 c.c.) volume of 2 

 per cent mercuric chloride solution containing 0.8 per cent hydrochloric acid. The 

 mixture is shaken vigorously in a stoppered flask and allowed to stand a few hours. 

 Centrifugate for 10 minutes, pour the supernatant liquid through a dry filter into a 

 graduated cylinder. An aliquot of the nitrate is taken, the mercury is removed 

 with hydrogen sulphide and the latter by a current of air. The liquid is exactly 

 neutralized and concentrated on the water-bath, or better, at 50 in a vacuum, 

 MgO added, and the mixture distilled in a vacuum at 45 to get rid of ammonia. 

 The volume should now be about 30 c.c. A little solid barium chloride and barium 

 hydroxide are added and 1.5 c.c. of 0.5 per cent solution of phenolphthalein. 

 Filter. Neutralize accurately to sensitive litmus paper. Add neutral formalin 

 solution and titrate with N/s NaOH as described in the chapter on quantitative 

 analysis of urine (page 504). 



6. Ammonia. Method of Folin and Denis. The determination' of ammonia in 

 blood is attended with considerable difficulty because of the fact that it is present 

 only in very small amounts, and to the fact that blood very readily and quickly 

 undergoes changes which are accompanied by the formation of ammonia from the 

 nitrogenous compounds present in the fresh blood. 



Of the methods proposed for its quantitative estimation that of Folin and Denis 2 

 is perhaps least unsatisfactory. 



Principle. The method is based upon the liberation of the ammonia from fresh 

 blood by aspiration after adding sodium carbonate solution, and the Nesslerization 

 of the solution into which the ammonia is aspirated. 



Procedure. Place 10 c.c. of systemic blood or 5 c.c. of portal or mesenteric blood 3 

 in a large Jena test-tube. Add 2 or 3 c.c. of a solution composed of 15 per cent 

 potassium oxalate and 10 per cent sodium carbonate, and about 5 c.c. of toluene. 

 Connect the tube for aspiration as in the method for urea in blood; start the air 

 current, and run as fast as the apparatus will stand for 20 to 30 minutes. Collect 

 the ammonia in another test-tube containing i c.c. of water and 5 or 6 drops of 

 tenth-normal acid. 



At the end of the aspiration Nesslerize in the usual manner (see method for non- 

 protein nitrogen, page 273) but more cautiously, adding in all not over i c.c. of the 

 previously diluted Nessler reagent (dilution 1:5). Transfer the contents of the 

 receiver to a 10 c.c. volumetric flask, fill to the mark with ammonia-free water, and 

 mix. Fill a 100 mm. polariscope tube with the mixture and close as for ordinary 

 polariscopic work. 



Prepare two standard solutions, one containing 0.5 mg. and the other i.o mg. 

 of nitrogen, these being Nesslerized simultaneously with the unknown solution and 

 made\ip to volume (100 c.c.). Use the standard possessing a tint most similar to 

 that of the unknown. Compare in a Duboscq colorimeter, using the ordinary 



1 Constantino: Bioch. Zeit., 55, 419, 1913. 



2 Folin and Denis: /. Biol. Chem., n, 532, 1915. 



3 The blood must be freshly drawn. For method of obtaining blood, see page 272. 



