AGRICULTURAL CHEMISTRY AGROTECHNY. 503 



ruainder of the liquid ; the extract made alkaline and heated on the water hath 

 until all the alcohol is driven off. then made up to a hulk of 50 cc, mixed 

 with 5 gm. sodium chlorid, acidified with dilute sulphuric acid, heated to the 

 boiling point, cooled, and filtered, the filtrate shaken with ether, and finally 

 washed several times with water. The etlier extract is evaporated to dryness 

 at a low heat. The extract thus obtained can be used for the tests, or it may 

 be converted into salicylic acid by the procedure recommended in the article. 



Comparative tests were made with different methods for detecting the 

 benzoic acid. For the test with ferric chlorid the residue was dissolved in 

 water containing ammonia, the solution evaporated to 1 cc, and 1 drop of 1 

 per cent ferric chlorid solution added. The author was able to detect 0.02 

 per cent of benzoic acid in smoked and fresh flesh and from O.OO.'j to 0.01 per 

 cent in fats. Rohrig's ethyl-ester odor test was sensitive to 0.02 per cent 

 benzoic acid in fresh meat and 0.01 per cent in fats, but no result was ob- 

 tained with smoked meat. Erevan's test gave negative results. Bj'- converting 

 the benzoic acid into benzaldehyde, 0.04 per cent was detected in fresh meat, 

 but with smoked meat the result was negative. With lard. 0.01 per cent could 

 be found, while the results with margarin were variable, and with butter 

 always negative. Mohler's test, which is based on the conversion of the ben- 

 zoic acid into the red-brown colored ammonium compound of amido-nitro or 

 the diamido-benzoic acid, also yielded negative results. 



Detection of benzoic acid in meats and fats, K. Fischek and O. Gruenebt 

 (Ztschr. Untersuch. Nahr. u. Genunsmtl., 29 {1910), No. 9, pp. 580-583). — The 

 authors investigated the von dor Heide and Jakob modification (E. S. R.. 23, 

 p. 12) of Mohler's method,^ which avoids the high temperature recommended 

 by Mohler. and found that it could also be employed for detecting benzoic acid 

 in butter, lard, and fresh, pickled, and smoked meats. 



The authors have also modified their own method, which is given in the 

 abstract above, for extracting the benzoic acid from butter and fats, and now 

 proceed as follows : Fifty gm. of the butter are placed in a 300 cc. Erlenmeyer 

 flask with 150 cc. of a 1 per cent sodium bicarbonate solution, boiling for 

 5 minutes, and shaking during the process. The hot fluid is transferred to 

 a separatory funnel, and the aqueous solution, which now contains the benzoic 

 acid, is separated from the fat. The cloudy alkaline solution is then neutral- 

 ized with half-normal sulphuric acid, using methyl orange as the indicator. 

 In order to obtain a clear solution quickly, 10 cc. of Fehling's copper sulphate 

 solution and 10 cc. of potassium hydrate (containing 31.15 gm. potassium 

 hydrate per liter) are added. After the pi-ecipitate has settled the supernatant 

 fluid must possess a slightly acid or neutral reaction. The solution is filtered, 

 the clear filtrate acidified with dilute sulphuric acid, and the benzoic acid 

 extracted with ether. The ether extract is washed several times with water 

 and the ether evaporated on the water bath with the aid of a current of air. 

 On detecting the benzoic acid it is converted into salicylic acid, according to 

 the method recommended by these authors. 



The authors have also investigated the Jonescu method (E. S. R., 21, p. 523), 

 but found it to be less valuable than the modified Mohler method. 



Estimation of tartaric acid, A. Heczko (Ztschr. Analyt. Cliem., 50 (1911), 

 No. 1, pp. 12-21; ahs. in Analyst, 36 (1911), No. 42i, p. 1//8).— In the hands of 

 some chemists the Goldenberg method,* which is the method usually employed, 



"Bui. Soc. Chim. Paris, .3. ser.. 3 (1890), p. 414; Ztschr. Analyt. Chem., 36 

 (1897), p. 202. 

 & Analyst. 33 (1908). p. 200. 



