﻿32 CLOVER. 



deemed sufficient to undertake the problem. Preliminary experiments 

 showed that a considerable proportion of a crystalline acid was formed 

 by oxidation with 4 per cent permanganate in the cold. The product 

 obtained appeared to be a mixture. 



CLASSIFICATION OF THE TERPENE OILS. 



(1) Of the 21 samples examined, 10 gave pure dextro-limonene. 



(2) Nine of the remaining oils, including all hut the two optically 

 inactive ones, contained more or less phellandrene. The nine phellan- 

 drene-containing oils fall into two classes. 



(a) Those showing an initial boiling point decidedly lower than that 

 of phellandrene and possessing a peculiar, indefinable odor. These are 

 the oils from Samples VII, X, and XV, and all are dextro-rotatory. 

 The oil from Sample X has the lowest initial boiling point and its rotation 

 is decidedly the smallest. The data obtained upon the physical constants 

 of these oils do not lead to any decided conclusion, but it appears ex- 

 tremely improbable that any body like pinene is present in them. Prom 

 the marked regularities which are found to exis't in the different samples, 

 it seems very probable that we have present' in these three cases one of the 

 unidentified terpenes of the limonene series. 



(&) Of the remaining six phellandrene-containing oils from Samples 

 V, XI, XIV, XVI, XVII, and XVIII, one was lsevo-rotatory and the 

 others, although they were all highly dextro-rotatory, showed a consider- 

 able variation in rotation. The two giving the greatest variation also had 

 a slightly higher boiling point, while the others, to judge from their 

 boiling point, appeared to be pure phellandrene. As has already been 

 brought out in the discussion of Sample XI, this oil consisted largely of 

 lsevo-limonene. 



In summing up this work it was suspected that the variation in the 

 rotation of the other phellandrene containing oils was due to the presence 

 in them of small amounts of lawo-linionene which could not be detected. 

 As shown in the experimental work which is detailed just below, the pres- 

 ence of lsevo-limonene could readily be proved by the addition of a small 

 portion of dextro-limonene to the oil and the subsequent isolation of 

 dipentene tetrabromide. The presence of leevo-limonene was prove 

 in oils XVII,A, purified, and XVIII.A, purified. XIV,A, purified, had 



been used up in previous work. In V, A, purified (a ■ D = -j-122°.6) 

 and XVI, A, purified (a ^=129°. 8), it can not be definitely concluded 

 whether or not la3vo-limonene was present and it can be readily seen 

 that if present at all it was in very small amounts. 



Detection of Icevo-Umonene. — A portion of XVII, A, purified, was added to 

 one-fifth its weight of dextro-limonene and 3 grams of the mixture treated 



