276 EXPERIMENT STATION KECORD. 



decant the solution into a 1,000 to 1,200 cc. flask, wash the residne thor- 

 oughly by decantation, and proceed as usual. The modified Kjeldahl 

 method is used only when a large amount of nitrates is present. 



The reporter was also instructed to test the method proposed by 

 Hilgard for the determination of nitrogen in humus, Avith a view to its 

 adoption bj- the Association. He was further instructed to undertake a 

 study of the acidity of soils. 



A paper on tlie action of different soil solvents by H. A. Huston and 

 J. M. Barrett was read by the former. 



Feeding fitxffs. — J. B. Lindsey submitted a report on cooperative tests, 

 by 6 chemists, of the Sachsse, Maercker, and diastase methods for starch 

 in potatoes and Buffalo gluten feed. 



The reporter held that it was useless for the Association to continue 

 studies of the so-called Weende method for the determination of the 

 proximate constituents of foods and feeding" stuff's, and that advance in 

 the study of such substances could be made only through efforts to 

 separate tbe different groups into their individual constituents. He, 

 therefore, selected for the work of the year the above methods for deter- 

 mining starch. 



The results of this investigation show that the Sachsse method as a 

 rule gave higher results than the others in case of impure j^roducts. 

 It was therefore decided that this method .should be adopted as official 

 only for potatoes and conmiercial starches. The other methods appeared 

 to be about equally convenient and accurate for geueral purposes. 



All of the methods adopted as provisional last year, except that for 

 pentosans, were made official. The following factors for calculating 

 pentosans and their products, based upon the work of Tollens, were 

 substituted for those adopted last year. 



Weight of hydrazoue X 0.516 + 0.0104 -f- amouut of substance taken ^= furfurol. 

 Furfurol X 1.84 = pentosans. 

 Furfurol X 1.65 = xylau. 

 Furfurol x 2.03 = araban. 

 Pentosans — 0.88 = pentoses. 



A paper by J. B. Lindsey and E. B. Holland on the " Distribution of 

 galactans in agricultural plants " was read by J. B. Lindsey. It was 

 shown that although galactan is quite generally present in plants, 

 being quite abundant in the leguminous plants, especially the clovers 

 and lupines, it occurs only in very minute quantities in the grains of 

 cereals and some other products. 



A paper on " The phlorogluciu method for the estimation of pen- 

 tosans," by J. B. Lindsey and E. B. Holland, was read by the former. 

 This method is described in detail, and comparative tests of it and of 

 the phenyl-hydrazin method on hay, grass, gluten feed, and lupine seeds 

 are reported. The 2 methods gave concordant results, except in the 

 case of English hay and branch grass. 



