KOSE GERANIUM OIL AND ITS SUBSTITUTES. 25 



indicate the number of milligrams of potassium hydroxid required 

 to neutralize the acidity of one gram of oil. The esters were esti- 

 mated by adding an excess of alcoholic potash to the above solution, 

 heating to boiling with a reflux condenser for about one hour, thru 

 titrating back the excess of alkali by means of decinormal acid. From 

 the amount of alkali consumed the necessary calculations can readily 

 be made, either as geranyl tiglinate (C 10 H 17 CO 2 C 4 H 7 ) or geranyl 

 acetate (C 10 H 17 CO 2 CH 3 ), as the case may require. The percentage of 

 alcohol, free or combined, was determined by acetyliziug a given 

 weight of oil with an equal amount of acetic anhydrid, in the 

 presence of fused sodium acetate. The acetylized product was 

 purified by washing with water, and rendered anhydrous by mcan^ of 

 fused sodium sulphate. A given weight of the acet} r lized oil was then 

 saponified, as outlined above for determining esters, and from the data 

 thus secured the desired calculations were made. 



In computing the amount of alcohol, both free and combined, in 

 geranium oil, it must be remembered that the chief ester of the natural 

 oil is a tiglinate, and on acetylizing with acetic anhydrid the free alco- 

 hols are converted into acetic esters. We therefore have a mixture 

 of esters on which to base our calculations. The percentages of 

 alcohols given in the table above were computed from the mixed esters. 

 The question might arise as to whether any of the tigjinic group was 

 replaced by the acetyl group, but this inquiry can not be answered 

 here. 



The first African oil, the second Algerian oil, and the Reunion oil are 

 normal in every respect. The African oil No. 2 and the Algerian oil 

 No. 1 can be considered normal except as to solubility, and it is doubt- 

 ful whether these oils can justty be considered adulterated. 'Flic third 

 African oil is low in alcohol content, high in optical rotation, and 

 insoluble in the proper amount of 70 per cent alcohol. The fourth 

 African oil is a spurious product, which did not respond affirmatively 

 fora hydroxyl group when tested by means of acetyl chlorid in the 

 conventional manner, thus indicating the absence of any alcohol. The 

 palma rosaoil and the Turkish oil No. 2 are both normal India products, 

 while the first Turkish oil is abnormal. Both of the ginger-grass oils 

 are entirely different from anything described in literature. Judging 

 from the high specific gravities and high optical rotations, these gin^vr 

 grass oils are not adulterated with either turpentine or mineral oil. 

 The last oil named, Rhodonol II, is apparently a fairly pure geraniol. 



It is clearly evident from the data obtained in this investigation and 

 other work on record that a scientific adulterator could readily mix 

 some of the cheaper geranium oils with the expensive, high-grade 

 products without much fear of detection by the analytical methods at 

 present available. Apparently the only satisfactory procedure at our 

 disposal for securing the proper quality of geranium oil is the use of 

 well-trained olfactories. 



