OF THE DIPTEROCARPACE Æ. 117 
found expedient to constitute 3 separate genera. Pierre, l. c. 
tab. 241, justly relies upon the structure of the seed for separating 
Synaptea from Vatica, but in this respect our knowledge is as yet 
imperfect. Vatica chinensis (Roxburghiana, Blume) has fleshy, 
plane-convex cotyledons, both bifid to base, the radicle lying 
between the two cotyledons. Vatica obscura, Trim., has a similar 
embryo, but, while in the case of V. Roxburghiana the radicle is 
as long as the embryo, the cotyledons being attached to it at the 
base of the seed, the radicle of V. obscura is only half the length 
of the embryo. A structure similar to V. Roxburghiana has 
the embryo of V. sarawakensis, Heim, V. Ridleyana, n. sp., and 
V. pedicellata, n. sp. ; while V. lanceefolia, Blume, and V. moluc- 
cana, Burck, more resemble V. obscura in that respect. V. Shu- 
manniana, Gilg, has similar cotyledons, but an exceedingly short 
radicle. All these species belong to the subgenus Retinodendron. 
The embryo of V. PAilastreana, Pierre, of the same subgenus, 
l. c. tab. 237, is different, and the same must be said of Reti- 
nodendron Scortechinii, King. Of Synaptea, S. faginea has thick 
fleshy cotyledons, both bifid to base, the inner enclosed by the 
outer, which is curved as in Shorea (seeds from Perak and 
Mergui) Pierre, tab. 242, had possibly unripe seed.  Pierre's 
description of S. astrotricha, l. c. tab. 240, agrees with what I 
have found in the case of S. faginea, except that he mentions 
* un reste d'albumen entre les cotylédons.” In the case of 
S. Dyeri he mentions a gelatinous albumen ; the embryo, however, 
is apparently similar to that of S. faginea and astrotricha. But, 
as said before, our knowledge of the seed of Retinodendron and 
Synaptea is as yet tooimperfect to give us safe generic characters 
for these two groups. 
A few words must now be said regarding the names of these 
three subgenera. King in his materials for a flora of the 
Malayan Peninsula (Journ. Asiat. Soc. Bengal, lxii. part 2, 
p. 100), and Trimen in his Handbook to the Flora of Ceylon, 
accept what has here been called Retinodendron and Synaptea as 
distinct genera. King, however, calls the latter Vatica, which 
name Trimen applies to the former genus. I am disposed to 
agree with Trimen. Vatica chinensis, Linn., as figured in J. E. 
Smith's Icones ineditz, tab. 36, is undoubtedly Vatica Roxburgh- 
iana, a tree of the subgenus Retinodendron, with short fruiting- 
calyx, the segments of which are equal. The name Synaptea 
(Sunaptea clearly is a misprint) was chosen by Griffith because 
his S. odorata (Vatica grandiflora, Dyer) has the calyx slightly 
