Eugenia in Hawaii — Wilson 
163 
Merrill and Perry (1938^, 1938^, 1939), in 
studies of Indo-Chinese, Chinese, and Bor- 
nean species of Eugenia , redefined the group, 
placing most of the species of those regions 
in the genus Syzygium. Geographically, their 
new concept limits Eugenia mainly to tropical 
America, and Syzygium primarily to the tropics 
of the Old World. 
The genus Eugenia in the Hawaiian Islands 
is represented by eight species as I interpret 
them. In my opinion, according to the Merrill 
and Perry system, four of the Hawaiian rep- 
resentatives of Eugenia would be classified in 
the segregate genus Syzygium , while four 
would fall in Eugenia sensu strictu. 
Eugenia malaccensis and E. Jambos have 
fruits which contain a rather large seed loosely 
attached within the pericarp. In both species 
the embryo has two distinct, fleshy cotyle- 
dons. The surface of the cotyledons is very 
rugose and the thick seed coat adheres firmly 
to it. Young seedlings of E. malaccensis were 
found growing which had the two cotyledons 
still attached. The pericarp had rotted away 
and no evidence of it could be seen, but on 
the surfaces of the cotyledons remains of the 
seed coat were still present. 
In fruits of Eugenia Cumini the fleshy peri- 
carp peels off, leaving the seed coat firmly 
adhering to the cotyledons. The cotyledons 
are fleshy and distinct. In dried, raw fruits the 
seed coat readily peels off with the pericarp. 
However, when boiled, the pericarp may be 
easily removed without disturbing the seed 
coat. 
Upon removal of the pericarp the testa of 
Eugenia sandwicensis also remains attached to 
the cotyledons. The two fleshy cotyledons 
are not consolidated. In boiled dried material, 
depending on the amount of care given it, the 
pericarp may or may not peel, away from the 
seed coat. 
The seeds of Eugenia rariflora and E. koolau- 
ensis have their cotyledons plainly separated, 
partly fused, or entirely consolidated. When 
the cotyledons are consolidated the line of 
their opposing faces is not evident. In both 
dried and fresh fruits the thin seed coat peels 
off with the pericarp. 
On the basis of the studies I have made, it 
seems evident that neither the character of the 
seed coat nor that of the embryo offers a 
satisfactory basis for the reclassification of the 
group. Neither one of the characters is con- 
stant. The degree of fusion of the cotyledons 
varies even within a single species. 
The conclusions I have drawn on the basis 
of investigations of the Hawaiian representa- 
tives of Eugenia strongly support Henderson’s 
rejection of the segregation of species of 
Eugenia into the genera Syzygium and Eugenia 
sensu strictu. 
No attempt was made to study the wood 
anatomy of the Hawaiian species of Eugenia. 
In view of Ingle and Dadswell’s findings 
(1953) I suggest that additional more exten- 
sive investigations may reveal other morpho- 
logical characters to support their conclusions. 
The characters of the seed, since they are so 
variable, do not justify the split. 
VARIATION AND DISTRIBUTION OF THE SPECIES 
IN HAWAII 
The genus is represented in Hawaii by only 
four indigenous species, three of which are 
endemic. Of these four, Eugenia rariflora and 
E. sandwicensis show the greatest degree of 
variation within the entire group. Both show 
variations in leaf size ranging from very small 
to rather large. 
The greatest degree of variation is seen in 
Eugenia sandwicensis. The forms of E. sand- 
wicensis with the larger leaves are found most 
often on Kauai; the forms with elliptical and 
elliptical-lanceolate leaves are more prevalent 
on Maui, and most of the Oahu representa- 
tives have the smaller obovate leaves. On 
Lanai the majority of the representatives of 
this species have small obovate leaves similar 
to those of the most common forms on Oahu. 
Most of the members on Molokai have leaves 
which resemble the forms on Maui, although 
in general they are not as long. It is possible 
