AGRICULTUKAL CHEMISTRY. 523 



The determination of small quantities of iron, W. McK. Marriott and 

 C. G. L. Wolf (Jour. Biol. Chem., J (1906), yo. 6, pp. 451-^61).— The method 

 proposed is based upon the color reaction which ferric salts show when treated 

 with an alkaline thiocyanate. Acetone is recommended as a solvent and de- 

 tailed directions are given for usins the method in determining the iron in 

 tissues and oriranic fluids, blood, urine, and inorj^auic substances. 



Determining the crude fiber content of cocoa, W. Ludwig (Zt.sc]ir. liitcr- 

 siich. yahr. ii. Gciiu-ssintl, 12 (IDOd), No. 3, pp. 1.53-15!)). — The crude fiber con- 

 tent of a number of samples of cocoa beans, cocoa shells, cocoa sprouts, and 

 commercial cocoa and chocolate were determined with a view to detectin.i; adul- 

 teration. Cocoa from which the fat was not extracted was found to contain o.Tl 

 to 4.42 per cent crude fiber, fat-free cocoa 4.98 to 5.96 per cent, and cocoa shells 

 14.47 per cent. The differences between the cocoa and cocoa shells are so great 

 that determining the crude fiber content may serve as a means for the detection 

 of undue amounts of shells in the ground cocoa. 



The estimation of crude fiber in cocoa goods, H. Mattiiks and F. Mullek 

 (Ztschr. riitersiich. Nalir. u. (IriiK.s.'iintl.. L> ( 1<>06), No. 3. pp. l.jH-Kil). — .V com- 

 parison of methods. 



Determining the digestible protein in feeding stuffs, A. Stutzer (.four. 

 jAindir.. J} {1006). No. 3. pp. 23.5-.>.'j<l. pJ. 1). — The determination of digestible 

 protein in feeding stuffs and related questions are critically discussed on the 

 basis of analytical and other data. 



In the author's opinion the separation of proteid from nonproteid compounds 

 by the aid of copper hydroxid is satisfactory, and he considers it immaterial 

 whether a solution of copper hydroxid is used or whether a copper salt is used 

 and the copper hydroxid formed during the analysis. In both cases care 

 should be taken to keep the licpiid slightly acid. This is done by using a few 

 cubic centimeters of strong alum solution when copper hydroxid is used as a 

 reagent and by using a little less sodium hydroxid than is necessary for the 

 complete change of copper sulphate into copper hydroxid when the copper salt 

 is used as a reagent. 



A simple method of separating proteids of feeding stuffs from peptones is 

 considered desirable since by the present method the peptones are not entirely 

 precipitated b.y copper. 



The pepsin soluble protein in feeding stuffs may be determined by treating for 

 48 hours at blood temperature 2 gm. of material with 500 cc. of gastric ferment 

 prepared according to the author's earlier directions or 250 cc. prepared by the 

 method descril>ed in the investigation ^'eported below, the acidity of the solution 

 being increased until it reaches 1 per cent of hydrochloric acid. 



Such artificial digestion differs from natural digestion and so the author 

 believes that the pepsin pancreatic method of determining digestibility should 

 receive further study. 



New experiments have shown that the results obtained by the pepsin method 

 are not always the same as those obtained by natural digestion. The need of 

 additional experiments on the nitrogenous metabolic products of the feces is 

 insisted upon. 



Studies on simplifying the method of estimating pepsin soluble constit- 

 uents of feeding stuffs, A. Stutzkr, II. \VA.N(iNicK, and W. Kothe {.Jour. 

 LdinJir.. :,', (lOOH). No. 3. pp. 265-272).— The conclusion is reached that the 

 pepsin solution used in determining the digestible protein of feeding stuffs 

 should be twice as strong as that conuuonly us(h1. and directions are given for 

 preparing such a reagent from fresh pig stomachs. 



