CHEMISTRY. 511 



_ ! 



sulphurous acid and boil gently to drive off excess of SO2. This treat- 

 ment reduces the chlorate to chlorid. Neutralize with pure calcium 

 carbonate and titrate for chlorin. The perchlorate is determined by 

 the method described above. 



Experiments in the determination of cell-v/all material, hemi- 

 celluloses, and cellulose in plants, A. Kt.kibkji {Lmidir. Vei's. 

 Stat, SJf. {1900), jVo. 3-4, j)jy. 161-31J).— The author reports a com- 

 parative study of the Henneberg (Weende), Lange (E. S. R., 8, 

 p. 741), Hofl'meister (E. S. R., 10, p. 606), and Schulze (E. S. K., 8, 

 p. 741) methods of determining crude fiber, on a variety of coarse 

 fodders, palm cake, etc. Concordant results could not be obtained by 

 the Lange method, duplicates differing b}^ as much as 60 per cent. 

 The author agrees with Suringar and Tollens (E. S. R., 8, p. 742) that 

 the cellulose is attacked to a considerable degree in this method. 

 Both the Hoffmeister and Schulze methods, which are similar in that 

 potasium chlorate is employed, gave crude fiber containing too much 

 protein. The results by the Schulze method were, however, more 

 concordant than by the Hoffmeister method. The author modified 

 these 2 methods by boiling for 2 hours with 1.25 per cent sulphuric 

 acid, obtaining materially lower results, indicating that the crude 

 fiber as determined by the original methods contains considerable 

 quantities of material easily soluble in dilute acids (hemicelluloses). 

 The results with this modification were almost invariably concordant. 

 The author believes that if the Hoffmeister or Schulze methods are to 

 be used at all they should be used with the modification proposed, 

 although he does not claim that the method is free from o))jecti()n, as 

 the cellulose obtained may not be true cellulose and may contain 

 oxycelhdose. 



An attempt Avas made to determine the total amount of cell-wall 

 constituents in plants by treating 3 gm. of air-dry substance with 

 200 CO. of 0.1.5 per cent potash solution for 2 to 3 days, collecting on a 

 dried and weighed filter, washing with cold and warm water, with cold 

 and hot alcohol, and finally with ether, and then drying to constant 

 weight. The ash and protein were determined and deducted. This 

 method and the use of alkali solution made l)y dissolving 5 gm. of 

 potassium hydrate in 1 liter of water gave practically the same results. 

 The end product by both methods contained quite large amounts of 

 protein. In the case of substances rich in starch the material was 

 treated with malt extract. The results of these determinations as 

 given by the author are not considered altogether reliable, being pos- 

 sibly too high in some cases and in others too low. The results were 

 considerably higher than those for crude fiber in the same materials. 



Experiments were also made in the determination of hemicelluloses 

 by treating the cell-wall material, as obtained above, with 1.25 per 

 cent sulphuric acid, but the question as to how long the treatment 

 13411— No. 6 2 



