410 EXPERIMENT STATION RECORD. 



mining starch in corn meal, lupine seed, and wheat middlings; (2) the 

 amounts of pentosans and galactans 1 in different feeding stuffs; and 

 (3) a comparison of the phenylhydrazin and phloroglucin methods for 

 determining pentosans. 



The reporter also tested the method proposed by Wiley for the deter- 

 mination of starch in comparison with the methods noted above. This 

 method is as follows : Extract the fat from 3 gm. of material, dry the 

 residue, and digest for 24 hours at 3i atmospheres in an autoclave 

 with J gm. salicylic acid dissolved in 50 cc. of water. After cooling 

 make up to 250 cc, filter, and heat 200 cc. of the nitrate in a 500 cc. 

 flask with 15 cc. of hydrochloric acid (sp. gr. 1.125) for 2£ hours at 

 100° C. Neutralize exactly with sodium carbonate, cool, make up to 

 500 cc, and determine dextrose in 25 cc. by Allihn's method. 



Both the Marcher and Wiley methods gave higher results than the 

 diastase method. The salicylic acid method gave lower results on 

 corn meal, the same on wheat middlings, and higher on lupine seed. 

 Examination of the residue left after treatment by the different methods 

 indicated that the high results were due to the conversion of the pen- 

 tosans and galactans. The amount of the latter present, therefore, has 

 an important influence on the accuracy of pressure methods of deter- 

 mining starch. The results of tests for pentosans and galactans in a 

 large number of feeding stuffs and farm products show that the latter 

 are not so widely distributed nor present in such large amounts as the 

 former. 



The difference between the nitrogen-free extract as usually deter- 

 mined and the sum of the actual determinations of the different carbo- 

 hydrates (using the diastase method for starch) was 8.74 per cent in 

 case of corn meal, 15.82 per cent in case of middlings, and 10.74 per 

 cent in case of lupine seed. 



The phloroglucin and phenylhydrazin methods 2 were compared by 

 5 chemists on the 3 substances named above. The former gave some- 

 what higher results as a rule. 



This report was discussed by H. W. Wiley, A. L. Winton, and others, 

 H. W. Wiley especially pointing out some sources of error in the 

 diastase method. 



B. W. Kilgore called attention to a study which he has undertaken 

 of methods of determining fat in dry and green fodders. 



J. L. Hills described briefly the mill used at the Vermont Station for 

 preparing samples of feeding stuffs for analysis. 



The diastase method was adopted for all substances except commer- 

 cial starches, but in carrying out the method the solution obtained by 

 treatment with diastase is to be heated with acid for 2i hours instead of 



'U. S. Dept. Agr., Division of Chemistry Bui. 49, pp. 49, 51; Massachusetts Hatch 

 Sta. Rpt. 1896, p. 92 (E. S. R., 9, p. 372). 



-U. S. Dept. Agr., Division of Chemistry Bui. 49, pp. 49, 53. Massachusetts Hatch 

 Sta. Rpt. 1896, p. 97 (E. S. R., 9, p. 322;. 



