AGRICULTURAL CHEMISTRY — AGROTECHNY. 411 



"When the automatic stopping of the vacuum distillation has taken place, the 

 flask is connected with a long condensex*, and a current of steam passed through 

 until the volume of the distillate reaches 200 cc. This distillation is so regu- 

 lated as to consume at least from 30 to 45 minutes. The volume of material 

 In the flask is kept as nearly constant as possible by heating with a flame. 

 When the steam distillation is carried out in less time than this, it is almost 

 invariably found that a sufficient amount of less readily volatile material is 

 driven over to produce murkiness and to interfere seriouslj' with the titration. 

 The same efCect is observed when the volume of oil with water in the flask 

 becomes tpo low. When a large volume of water accumulates in the flask, the 

 results appear to be too low, due to incomplete volatilization of the ester. The 

 steam distillates are always found to be slightly acid to phenolphthalein, but no 

 relation between this acidity and the saponification value has been observed. 



" A concentrated aqueous solution of the theoretical quantity, or an excess 

 of semicarbazid hydrochlorid with an equivalent amount of crystalline sodium 

 acetate, is now added. This is calculated from the aldehyde content, deter- 

 mined previously by Hiltner's method.'^ When an insufficient amount is used, the 

 end point is not sharp. One hundred cc. of 95 per cent alcohol is then added, the 

 mixture shaken around for a few minutes, and allowed to stand for from 

 10 to 15 minutes, or longer if convenient. A large bulk of citral semicarbazone 

 usually separates at this point. The solution is then neutralized to phenol- 

 phthalein, 50 cc. half-normal alcoholic KOH added, and the solution boiled xmder 

 a reflux for 2 hours. At the end of this time it is cooled to room temperature 

 without delay, under tap water, and the excess alkali titrated with half-normal 

 hydrochloric acid. It is necessary to use a much larger quantity of phenol- 

 phthalein than In ordinary titrations. Using a 100-gm. sample, the number of 

 cubic centimeters of half-normal alkali consumed, multiplied by 0.098 (the 

 value in grams of 1 cc. half-normal linalyl acetate), and an empirical factor 

 [1.28] gives the percentage of saponifiable matter present, calculated as linalyl 

 acetate." 



When applied to lemon extracts the procedure is as follows : 



" Four hundred gm, is distilled slowly from an ordinary side-neck flask until 

 the volume is reduced to from 50 to 75 cc. Steam is then passed through 

 until no more volatile oil comes over. The combined distillates are then treated 

 exactly as the steam distillate in the case of lemon oils, calculating the neces- 

 sary amount of semicarbazid from the citral value (1 gm. citral requires about 

 0.75 gm. semicarbazid hydrochlorid)." 



Experimental data obtained from mixtures of known composition indicated 

 that the method is quite accurate. 



A disturbing factor in Barfoed's test, W. H. Welkek {Jour. Amer. Chem. 

 Soc, 37 {1915), No. 9, pp. 2221-2230) .—In using Barfoed's test* to determine 

 the hydrolytic products of the action of hydrochloric acid on starch, the author 

 found it impossible to produce the typical glucose reaction but obtained a 

 greenish-white precipitate. The sodium chlorid present, formed by the neu- 

 tralization of the free hydrochloric acid with sodium hydroxid, seemed to be 

 the interfering substance. A greenish-white precipitate was formed instead of 

 the red cuprous oxid of the typical Barfoed's test when a solution of sodium 

 chlorid was boiled with Barfoed's reagent, or when sodium chlorid was added 

 to a solution of pure glucose and Barfoed's test was applied to the mixed 

 solution. 



"U. S. Dept. Agr., Bur. Chem. Bui. 132 (1910), p. 102; Jour. Indus, and Engin. Chem., 

 1 (1909), No. 12, pp. 79S-S00. 

 "Jour. Biol. Chem., 6 (1909), pp. XXXIII-XXXIV. 



