— 49 — 

 Analysis of the silver salt of the acetic acid boiling at about 120°: 



0,2729 g. subst: 0,1758 g. Ag. 

 Found Calc. for C 2 H 3 Ag0 2 



Ag 64,42 °/o 64,67 °/o 



The acid mixture which was isolated from the saponification liquors 

 of the ester fractions which had served to afford the test for the presence 

 of citronellol, as well as from the higher boiling oil fractions, contained 

 in those of its portions which boiled above 130° an acid of which the 

 b. p. was about 65° (7 mm.). At ordinary pressure this acid passed over 

 constant at 163°; and by its odour likewise it was recognised as w-butyric 

 acid. It solidified into crystals when placed in a freezing mixture. Its chloride 

 boiled between 35 and 36° (7 mm.) and, with ammonia, was converted 

 into w-butyric amide, m. p. 114 to 115°. Valeric acid, which is said to 

 occur in citronella oil, could not be isolated up to the present. The 

 distillation-residue of the acid mixture which remained behind was a solid 

 yellowish acid, crystallising from water in the form of needles, m. p. about 

 180°; this acid was not identical with veratric acid. 



Those portions from the preparation of citronellol which gave no 

 reaction with phthalic anhydride, and of which the b. p. lay between 85 

 and 90° (4 mm.), contained a hydrocarbon with an odour resembling that 

 of petroleum. By treating this substance with benzoyl chloride and pyridin 

 its alcoholic constituents were eliminated. The body, when heated in 

 vacuo with sodium, possessed the following properties: b. p. 95 to 96° 

 (5 mm.), b. p. 249 to 252°, d 15 o 0,8924, « D — 13° 2\ n D 1,49281. Combustion 

 showed it to consist of a sesquiterpene Ci 5 H 2 4, probably contaminated 

 with a slight proportion of methyleugenol, which is well-known to be the 

 principal constituent of the high-boiling fractions of citronella oil. No 

 derivatives such as a nitroso chloride, nitrite, nitrosate or hydrochloride 

 were obtainable. When the body was brominated in chloroform solution 

 there was a strong evolution of hydrobromic acid. In view of the dif- 

 ference in the constants it is questionable whether this body is identical 

 with the light sesquiterpene of citronella oil of which the presence was 

 detected by us several years ago 1 ). We have not gone into a further exa- 

 mination of the fractions, which consist principally of methyleugenol and 

 of the dextrorotatory sesquiterpene of which the presence has also already 

 been established on a previous occasion. 



• The examination detailed above has therefore shown that Ceylon 

 citronella oil contains the following bodies in addition to those previously 

 known:, a hydrocarbon of low specific gravity (terpene?), a body related 

 to linalool, an alcohol which may be found to be identical with thujyl 



') Report October 1899, 23. 



