46 METHODS OF QUANTITATIVE ANALYSIS. 



supports a smaller sheet-iron plate. Around the edge of 

 this plate six semi-circular notches are cut, equally dis- 

 tant from one another. The necks of the flasks fit into 

 these notches and are thus supported in an oblique po- 

 sition. For the heating, six Bunsen burners are fitted at 

 proper intervals to a gas pipe bent in the form of a circle. 

 Each burner is controlled by a stopcock. For the dis- 

 tillations a simple iron stand with sheet iron top is used. 

 Six holes are cut through and covered with wire gauze 

 to support the flasks. Bunsen burners fixed to a main 

 gas pipe supply the heat. The condenser consists of a 

 large rectangular tank, supported on a suital;>le stand and 

 provided with an inlet and outlet for cold water. 

 Through the tank are put six wide glass condensing 

 tubes, the connections being made with tightly fitting 

 rubber stoppers. 



B. METHODS OF PRECIPITATION OF ALBUMIN- 

 OIDS. 



In technical analyses and market control tests, where 

 it is unnecessary to determine the exact distribution of 

 the nitrogen in various combinations, the method of 

 Ivjeldahl's for the estimation of total nitrogen has 

 proved most satisfactory and reliable. For the 

 direct determination of albuminoid bodies a large 

 number of processes have been proposed. (Alco- 

 hol precipitation ; precipitation with tannin, pre- 

 cipitation with acetic acid according to Hoeppe- 

 Seyler, and with copper sulphate and sodium hy- 

 drate according to Ritthausen.) J. Munk has recently 

 made a study of these methods and has suggested sev- 

 eral modifications which tend to simplify the processes 

 and render them more accurate. Only those methods 

 which have proved to yield satisfactory results and which 

 are applicable to all sorts of milk will be considered here. 

 These methods all depend upon the complete precipita- 



