GEBANIACE.ZE 491
Ibs. of the plant, consisting of the entire stems with leaves and flowers,,
yielded 0'044 per cent, of oil having the following characters :—Specific gravity at 16°
Optical rotation
Refractive index at 18°
Saponification value
(after acetyltion0-888
- 5-39°
1-4600
96-25
226-08
Total geraniol 74-79 per cent.
Combined geraniol 28-19 „
Free „ 46-60 „
Esters as geranyl tiglate ...... 40-12 ,,According to Furukawa,
1
three varieties of geranium, Pelargonium
graveolens, P. radula, and P. denticulatum, have been cultivated on the
Oi experimental farm near Tokyo. Of these only P. denticulatum re-
sembled closely the geranium oil of commerce. This had a sp. gr.
0-886 at 20° C.; acid value, I'll \ ester value, 12-8; free alcohol as
geraniol 60 per cent.; total geraniol, 635 per cent. The yield of oil
was 5 per cent. The oil of P. graveolens had a sp. gr. 0-9178 at 20° C.;
<XD + 4° 6'; acid value, 183'6 ; ester value, 48 ; total geraniol, 221 per
cent.; free alcohol (as geraniol), 22 per cent. The yield of oil was from
175 to 2 per cent. The oil was found to contain a comparatively
large amount of an organic acid, identified as dextro-citronellic acid.
Laevo-menthone and laevo-citronellol were also isolated, by fractional
distillation. P. radula yielded 1*5 per cent, of oil, having a sp. gr.
0-9234; aD + 3° ; acid value, 5'6; ester value, 31'7 ; total geraniol, 26'3
per cent.; free alcohol (as geraniol), 17'26 per cent.
Geranium oil is soluble in 2'5 to 3 volumes of 70 per cent,
alcohol.
The relative proportions of the two alcohols, geraniol and citronellol,
in oil of geranium oil is a matter of considerable interest, and the de-
termination is frequently of importance. The only method available in
practice for this purpose is the combination of the acetylation process,
which gives the measure of the total alcohols and the formylation pro-
cess, by which the geraniol is decomposed, and the citronelloL
estimated.
This method has been referred to under Otto of Eose, but has been
more critically studied by Simmons
2
in reference to geranium oil.
Simmons has shown that the determination of citronellol in the presence
of geraniol by the formylation process is not strictly accurate, but the
results are exceedingly useful for comparative purposes.
The formylation process, as recommended by MM. Jeancard and
Satie, consists in heating 10 c.c. of oil with 20 c.c. of 98 to 100 per cent.
formic acid, in a flask attached to a reflux condenser for one hour on a
sand-bath. It may be mentioned that the addition of 2 grams anhy-
drous sodium formate per 10 c.c. oil, while it enables the mixture to be
boiled steadily on a sand-bath, does not affect the result.
To prove whether the results obtained represent the true citronellol-
content Simmons determined the " total alcohols " by acetylation, and
the " citronellol " by formylation of (1) pure geraniol, (2) pure citronellol,
and (3) a mixture of (1) and (2) in equal proportions, with the following
results:—
(^1) Jour. Ghem. Ind. Tokyo (1919), 22, 83.
2
Year Book of Pharmacy (1913), 565.