Kawa Constituents — Hansel 
295 
OCH 3 
OCH- 
CH = CH 
kawain 
OH 
1 H 2 ,(Pd) 
cr^o 
di hy d r o k aw a i n (marindin) 
OCH 3 
^J^"CH-CH - CH =CH - C~ CH“C0 2 H 
H 2 ,(Pd) 
OH 
OCH 3 
1 
CH 2 “CH 2 ~CH=CH-C = CH-C0 2 H 
H + 
kawaic acid 
-C0 2 
- CHoOH 
rA 
\ 
nV£ 
OCH 3 
1 
dihydrokawaic acid 
H + 
^”CH 2 -CH 2 “CH 2 -CH 2 -C=CH-C0 2 H 
tetrahydrokawaic acid 
0 
II 
ch=ch-ch=ch-c-ch 3 
cinnamal acetone 
-C0 2 
- CH 3 OH 
2 H 
2j(Pd) 
N 
ch 2 -ch 2 -ch=ch-c-ch 3 
6 -phenyl -3 - hexen - 2 -one 
1 H 2 ,( Pd) 
HOBr 
^ A-ch 2 -ch 2 -ch 2 -ch 2 -co 2 h 
pheny Ivaleric acid 
0 
II 
CH 2 “CH 2 -CH 2 -CH 2 -C-CH 3 
6 -phenyl-hexan- 2 -one 
Fig. 2. Chemical reactions of kawain and methysticin (only kawain is shown). 
(b) Nothing very remarkable can be said 
regarding the substituents on the benzene ring. 
In addition to unsubstituted derivatives we find 
the corresponding p-methoxy derivatives, dime- 
thoxy derivatives, and dioxymethylene com- 
pounds. No lactones occur in kawa which have 
a free hydroxyl group to which sugars could be 
linked to form glycosides. Lactones with a free 
or a glycosidically linked hydroxy group there- 
fore do not occur. Of course we have no answer 
to the question why the kawa plant does not 
produce free phenols or glycosides. This may 
have to do with the excretion cells of the plant. 
A particular constituent, in order to be de- 
posited in the excretion cells of the Piperaceae, 
must have a certain lipid solubility, a phenom- 
enon which is well known from the constitu- 
ents of essential oils. Substances with a free 
hydroxy group, or even with glycosidically 
linked hydroxy groups, obviously do not have 
a suitable partition coefficient for deposition in 
the excretion spaces. Of course, the correlation 
may also be reversed: since the plant synthesizes 
many lipophilic end products, excretion cells are 
formed. Be that as it may, a correlation no doubt 
exists. 
If we now combine the two possibilities for 
variation, (i) the degree of hydrogenation, that 
is, the number of non-aromatic double bonds, 
and (ii) benzene substitution, we arrive at a 
possible total of twelve structural variants 
(Table 1). Six of these have been known for 
a long time. We have worked out effective ana- 
lytical separation and testing methods which will 
allow us to find additional lactones which might 
occur in the plant in trace amounts. We have 
synthesized all twelve structural variants and a 
few additional ones. Furthermore we have in- 
vestigated kawa from various parts of Oceania, 
using samples from the island of Hawaii, the 
Fiji Islands, and from Samoa. As a result we 
now have the following picture. In addition to 
the six already known lactones we have demon- 
strated the presence of three additional deriva- 
tives so that today nine of a total of twelve 
pyrones are known as constituents of Piper 
