( 743 ) 



XXXIII. — I. On the Estimation of Carbon in Organic Substances by the Kjeldahl Method. 

 II. Its Application to the Atialysis of Potable Waters. By Charles Hunter 

 Stewart, D.Sc, M.B.* (From the Public Health Laboratory of the University 

 of Edinburgh.) (With Two Plates.) 



An easy and yet accurate method of determining carbon and nitrogen in organic 

 substances has long been a desideratum, especially among those engaged in the applica- 

 tion of chemistry to biological, hygienic, and agricultural questions. For the deter- 

 mination of nitrogen the method of Dumas, with its numerous modifications, is still the 

 only one applicable in all cases, but the time required for it, and the manipulative 

 dexterity necessary, has prevented its wide application for the above-named purposes. 

 The method of Will and Varrentrap, though less generally applicable, is easier, and, until 

 the publication of Kjeldahl's t method, was most frequently used in applied chemistry. 

 Kjeldahl claims for his method the same applicability and as great accuracy as the Will 

 and Varrentrap method, with the added advantage of greater ease in working. 



The Kjeldahl method is founded on the fact that most nitrogenous organic substances, 

 when heated along with strong sulphuric acid, are decomposed, their nitrogen being 

 converted into ammonia, which, in presence of the sulphuric acid, forms sulphate of 

 ammonia. The modus operandi given by him is as follows. The weighed quantity of 

 the substance (which may vary from O'l gramme to 07 gramme according to the 

 percentage of nitrogen) is introduced into a long-necked bohemian glass flask of about 

 100 c.c. capacity, and 10 c.c. of strong sulphuric acid added. The flask is now supported 

 in an inclined position on the ring of a retort stand covered with wire gauze and heated 

 by a low flame. After the action has been well established, the flame is raised until the 

 acid begins to boil. The heating is continued till the colour of the acid is pale yellow. 

 The time required varies with the substance, but in the majority of cases two to three 

 hours is sufficient. The flame is now removed and the flask and its contents allowed to 

 cool. When perfectly cool, a few grains of powdered permanganate of potassium is 

 dusted in and the flask again heated with a low flame for five or ten minutes. When the 

 colour of the acid has become green the reaction is ended. There is no loss of ammonia 

 in the process, either in the original heating with sulphuric acid or in the subsequent 

 treatment with permanganate of potassium, if carried out in the way described. A 

 known quantity of sulphate of ammonia, heated with strong sulphuric acid and subse- 

 quently treated with permanganate of potassium, gave, on neutralisation and distillation, the 

 amount of ammonia used. Kjeldahl points out that simple heating with sulphuric acid 

 for two to three hours will convert from 90 per cent, to 100 per cent, of the nitrogen 



* Thesis for degree of D.Sc. (Depart, of Chemistry), University of Edinburgh, 

 t Zeitschrift fiir Analytische Ghewie, Band 22 (1883), p. 366, &c. 



VOL. XXXVII. PART IV. (NO. 33). 5 X 



