— 92 — 
is identic with tanacetone dicarbonic acid, Semmler concludes tliat 
the following formulae must belong to the sabinene ketone and sabinene: 
(CHg), CH 
C 
H.,C 
CO 
Sabinene ketone 
HC 
CrCH^ 
Sabinene 
It is a remarkable fact that if the semicarbazone of sabinene ketone 
is heated with dilute sulphuric acid, the ketone is not regenerated, but 
an unsaturated hydrocarbon C9 4 of the boiling point 165 to 166° and 
specific gravity 0,836 at 20° is formed, to which the author attributes 
a penta-ring wdth two double linkages, and that with the limonene-type 
of the hexa-ring. 
It appears that in reducing sabinene ketone, not only the alco- 
hol C9H;^gO, but also the saturated alcohol CQH^^gO is formed, so. 
that the effect of the tri-ring which is near to the ketone group, appears 
to be analogous to that due to the presence of a double linkage. 
|AlcoholsJ 
Thujyl alcohol. While engaged in the study of the isomeri- 
sation of secondary into tertiary hydroaromatic alcohols, Kondakow^) 
observed that isomerisation also takes place with thujyl alcohol when 
it is converted into the chloride by means of phosphorus pentachloride 
or hydrochloric acid. If thujyl alcohol is converted, by means of 
phosphorus pentachloride, into thujyl chloride, the latter shows neither 
a constant boiling point, nor a uniform composition, and it is more- 
over not as indifferent as might be assumed after Semmler's state- 
ments. This is due to the fact, demonstrated by Kondakow, that 
the crude chloride is not a uniform compound, but on the contrary 
a mixture of two chlorides, w^hich probably consists chiefly of a ter- 
tiary compound. When treated with moist silver oxide, it yields 
a probably tertiary alcohol with a mouldy odour; treatment with alco- 
holic potash solution produces (if only with difficulty) a mixture of at 
least two hydrocarbons, which are closely allied in their properties, 
^) Chemiker-ZeiUmg 26 (1902), 720. 
