12 EXPERIMENT STATION BECOED. [Vol. 37 



The carbon was then burned out as completely as possible, again thorotighly 

 extracted with water, and the residue finally ignited and weighed as insoluble 

 ash. 



For the analysis proper 100 gm. of the ground straw was taken and treated 

 as noted above. The charred mass was extracted, however, with hydrochloric 

 acid and after filtration the carbon burned out as far as po.ssible, and the resi- 

 due again extracted with hydrochloric acid and then ignited and considered 

 as pure silica. The hydrochloric acid extracts were evaporated to dryness to 

 separate the soluble silicic acid, and the weight of the silicic acid thus obtaine<l 

 added to the insoluble silica to obtain the total silica. The hydrochloric acid 

 solution was made up to 250 cc. and aliquots taken for further determinations 

 by the procedures used for the wheat grain. The phosphorus was determined 

 in a separate 25 gm. sample which was treated with nitric acid and mag- 

 nesium oxid. The thoronp;hly burned mass was dissolved in nitric acid and a 

 few cubic centimeters of concentrated sulphuric acid adde<l, and evaporated 

 until sulphuric acid fumes were given off, after which the dish was placed in 

 a hot-air oven and kept at 160° C, or higher, to allow the separation of the 

 silicic acid. The phosjjhoric acid was then determine<l by the usual procedure 

 of precipitation by ammonium molybdate, and finally weighed as magnesium 

 pyroiihosphate. If necessary, the silicic acid could be removed by hydrofluoric 

 acid before the addition of the magnesium oxid. 



Tlie sulphur was determined in the same manner as In the wheat grain. 

 The same procedure was also used for the determination of manganese, except 

 that the silica wjis removed by concentrated hydrofluoric acid. 



A new qualitative test and colorimetric method for the estimation of 

 vanillin, C. Estes (Jovr. Jndvii. and Engin. Chem., 9 {1911), No. 2, pp. IJ^ 

 llflt). — After some preliminary experimentation the author devised a procedure 

 for the determination of vanillin which depends on the production of a violet 

 to violet-red color when a solution of vanillin Is treated with an acid solution 

 of mercuric nitrate. 



The procedure as applied to alcoholic extracts Is as follows: To 5 cc. of the 

 extract In a 50-cc. graduated flask, 6 cc. of water and 1.5 cc. of the acid mercuric 

 nitrate reagent are added. The standard is made up at the same time with 

 5 cc. of the standard vanillin solution, 0.5 cc. of the reagent, and 6 cc. of water. 

 The two flasks are placed in boiling water for 20 minutes. They are then 

 taken out, rapidly cooled, made up to the mark, filtered, and the colors 

 compared. 



For nonalcoholic extracts the standard Is prepared in the same way as for 

 alcoholic extracts, and the only change in the preparation of the unknown is 

 the quantity of reagent used, 1 cc. having been found to give the maximum 

 intensity of color. 



Analytical data .submitted indicate that the results obtaUied on commercial 

 vanilla extracts agree very closely with those from the gravimetric procedure. 



Summary of the composition of the wines of current consumption, G. 

 FiLAUDEATJ (Ann. Falsif., 9 (1916), No. 96, pp. SJfl-413).— These pages contain 

 detailed tabular analytical data of the various wines of the harvest of 1915, 

 together with comments on the data. 



Adsorption of tin by proteins and its relation to the solution of tin by 

 canned foods, B. C. Goss {Jour. Indus, and Engin. Chem., 9 {1911), No. 2. pp. 

 lii-148). — The work reported shows that " the solution of tin by canned foods 

 Is neither dependent upon, nor proportional to, the acidity alone, and, also, 

 that in the foods of relatively slight acidity which dissolve large amounts of tin 

 the greater part of the tin is in the form of an insoluble and stable complex." 

 The data Indicate tliat the phenomena are due to adsorption, and that the tin. 



