124 Flora of the Malayan Peninsula. 
unequal. Stamensin 3 flowers normally 9 or 12, sometimes more, 
rarely less (in one species 6), in four rows, those of the two outer 
rows with eglandular or occasionally 1—2-glandular filaments, those 
of the 3rd row with biglandular filaments, as are those of the 4th 
row when present, all the stamens with 4 cells, two cells introrse or 
the lower pair lateral; ovary 0 or rudimentary with rudimentary 
style and stigma, very rarely l-ovuled. Staminodes in 92 flowers 
normally 9 or 12, etc., as in 3, those of the outer 2 rows usually 
clavate or spatulate, those of the inner rows usually shorter subulate 
or clavate with a pair of glands at their bases; ovary enclosed in 
the perianth-tube or free, globose or ovoid; style usually thick, 
often curved ; stigma dilated, irregularly lobed. /rwit globose, ovoid, 
ellipsoid or cylindric, resting on the more or less enlarged perianth- 
tube and supported by the more or less enlarged pedicel; the 
enlarged perianth-tube sometimes very small and merely supporting 
the fruit, sometimes obconic or cupular partly enclosing it, some- 
times even much enlarged and almost wholly enclosing it; perianth- 
lobes usually deciduous, occasionally persistent; pericarp usually 
succulent. Seed 1; testa thin; cotyledons conform to the seed, 
fleshy.—Distris. Tropical and subtropical Asia, in India, Burma, 
the Malayan Peninsula and Archipelago eastwards to Japan, south 
to the Pacific Islands, Australia and New Zealand; scarce in Africa 
and America: species from 150 to 200, 
In preparing the following key, I have reluctantly been obliged to 
abandon the idea of arranging the species under the old subgenera having 
found the task almost impossible. However, in avoiding characters taken 
from the flowers and relying chiefly upon those of the leaves, inflorescence 
and fruit, I think I may have made it more easy to follow in the field, where 
a worker can only very rarely have at his command specimens of both ¢ and 
¢ plants as well as of the fruit. It may not be out of place to sound a note 
of warning as to the fruit; the enlarged perianth forming the cupule varies 
greatly as it develops gradually to maturity; and mature specimens are quite 
necessary for proper identification. The flowers in this genus are very variable 
but in a single umbellule there are usually one or two flowers with the normal 
number of parts, the rest frequently varying considerably from them, the 
central flower often having more than the normal number. 
I need say little on the reasons for the adoption of Litsea, Lamk., as the 
generic name; for the subject was fully gone into first by Jussieu in the Annales 
d’ Hist. Nat. VI. 197 who fixed on Litsea, and by Blume in Mus. Bot. Lugd.- 
Bat. I. 371 who decided to adopt Tetranthera, Jacq., and afterwards by 
Bentham and Hooker in the Gen. Pl. III. 162 who showed that Litsea, Lamk., 
was the correct name. They have been followed in this by Pax in Engler and 
Prantl Natur]. Pflanzenfam. IIT, 2, 199. 
