— 77 — 



normal potash liquor required for 30 cc. of the filtrate is calculated. For 

 the purpose of estimating the value of the phenylhydrazine solution a 

 blank test without oil is made in a similar manner. If it is found that 

 the quantity of decinormal potash liquor used up for 30 cc, of the filtrate 

 = a in the first, and = b in the second experiment, it follows that the 

 quantity of citral present in the amount of oil under test (s grms.) equals 

 a — b cc. of decinormal potash liquor. Hence, 1 cc. decinormal potash 

 liquor being equal to 0,0152 g citral, the percentage proportion of citral in 

 the oil is expressed by the following formula 



(a --•&)• 1,52 

 s 



The object of the extraction with benzene (in which we have followed 

 a suggestion of Kleber) is to re-clarify the solution, which becomes turbid 

 after the addition of the hydrochloric acid. We have observed that when 

 this has been done it is easier to recognise the change in the colour 

 during the titration. 



We summarise below the results which we have obtained by this 

 method with mixtures of known citral-content. These figures clearly show 

 the usefulness of the method. 



Found . 



Percentage of the solution 



2°/o 



2,4<>/o 



2,0% 



40/0 



4,1 % 



4,0 °/o 

 3,9% 



6 0/0 



6,2 0/0 

 5,9% 

 6,0 0/0 



7,3° 



6,9 0/0 

 7,0o/ 

 7,1 0/0 



From a paper by Chace ! ) we observe that several American chemists 

 have also obtained satisfactory results with Kleber's process. Comparative 

 determinations by the methods of Chace (Report April 1907, 116), Hiltner 

 (Report October 1910, 163), A. H. Bennett (Report April 1909, 110, October 

 1909, 153) and Kleber have shown that the last-mentioned method not 

 only appears to afford the best results, but is also the easiest to carry 

 out. The least satisfactory of all were the colorimetric tests; Hiltner's 

 method regularly gave results below the actual values, Chace's, above 

 them. In respect of accuracy of results Bennett's method came next to 

 Kleber's, but Bennett's method failed partly in this particular that not 

 |all the chemists who applied it were able to recognise beyond a doubt 

 the concluding point of the reaction. 



As a result of the experiences set forth above the official chemists 

 of the Bureau of Agriculture of the United States, at their 27 th Annual 

 Meeting, agreed to recommend Kleber's method for the determination 



*) U. S. Dep. of Agriculture, Bureau of Chemistry. Bulletin No. 137 (20. VII. 1911), p. 64. 



