CONVENTION OF OFFICIAL AGRICULTURAL CHEMISTS. 509 



The referee for the next year was instructed to test the so-called 

 international method (treatment with 3 per cent nitric acid) for deter- 

 mining- available potash, and Hallemann's method (treatment with car- 

 bonated water) for determining lime, and also to further test the 

 ammonium chlorid method for available potash. 



B. L. Hartwell, associate referee, reported the results of comparative 

 tests of methods by 3 analysts on hay ash. The results show that the 

 accuracy of the methods is largely dependent upon the soluble silica 

 present. The thorough revision of the methods was proposed, and it 

 was suggested that they be tested on mixtures of known composition 

 approximating the composition of ash. 



FOODS AND FEEDING STUFFS. 



The report on this subject, which was submitted by W. H. Krug, 

 referee, gave the results of comparative tests by 13 analysts of the 

 diastase method for carbohydrates in barley and bran, by 7 analysts of 

 the phenylhydrazin and phloroglucin methods for pentosans in the same 

 substances, and by 6 analysts of methods of determining galactan. 

 The importance of making a microscopic examination for the purpose 

 of showing when the conversion of starch by the diastase is complete 

 was brought out in these tests. The phloroglucin method for pentosans 

 as a rule gave higher results than the phenylhydrazin method. 



II. \V. AViley spoke of the difficulty of the exact determination of 

 starch, but stated that in his opinion the diastase method is the best 

 yet offered. He also stated it as his belief that our present methods 

 account for all substances which occur in the nitrogen-free extract of 

 cereals and cereal products. H. Snyder spoke of the importance of 

 determining the acids in food. H. A. Huston suggested that a study 

 of the alkali extract of foods and feeding stuffs would throw some light 

 on the constitution of the nitrogen-free extract. 



The following recommendations of the referee were adopted by the 

 association : Three grams of material are to be used in the determi- 

 nation of starch by the diastase method. The substance is to be 

 treated with diastase until the presence of starch is no longer shown 

 by the microscope. The acid solution after inversion is to be nearly 

 instead of exactly neutralized. The phloroglucin precipitate is dried 

 by aspirating h hour and then heating in an oven for 3i to 1 hours. 

 In the diastase method blank determinations to test reagents should 

 be made. In the determination of galactan the material for the test, 

 3 gin., is to be extracted 5 times with 10 cc. portions of ether. The 

 association voted to adopt the factors for calculating pentosans recently 

 proposed by Krtiger and Tollens 1 instead of those used at present in 

 the official method. The referee called attention to the importance of 

 using phloroglucin free from diresorcin. To detect this substance dis- 

 solve the phloroglucin in acetic acid and add a few drops of strong 

 sulphuric acid. A violet color shows the presence of diresorcin. 



ZtscLr. Angew. Chem., 1896, p. 33. 



