50 ANXALS OP THE EOYAL BOTABIO GABEES, CA10ÜTTA.
Ifalays on ncoount of the fragi-anco of its flowers. Those wUd amanga, Riimphms calls
Om'vigoe syîmtres, and of them he distinguishes throe sorts: —
1. Cmamja sylnestrh prima sivs h-îpliata (Malaiee 0dm),
2. CmmKja stjUestria secunda siva (mijnstifolia.
3. Cunmga sylvesiris iertia sive laiifoUa.
Of the first two Kumphius giws figures on t. 06 of the same Tohnne ; and, judging
from those figures, the plants woidt! fall into the modem genus FohiaUhia.
Linnious' Sfecm Pkmtamm was published in 17Ó3; therefore Rumphius' names are
in point of time, as they are in point of form, pre-Linnasan. Linnasns does not
accept 0,m,m,ja as a genns, and he refers to the Canaula of Emnphius <mly in a note
rmder Vvaria ZeaUmka. And the first botanists to adopt the Cawmn« of Enmphms as a
genns are Hook. fil. & Thomson (in Fl. Ind. 130). But in 1775 Aublet (in his
Ilistoir, des Flantis de la Ouiam Franeam) published, in regular Lmnajan fashion, the
.^enus Canatiga for the reception of a single species named 0. oaregow, of which he
TOve a figure (t. 2 « ) . Kiueteon years later (1794) Ruiz and Pavon (in their Froirmmn
Flm-oe Femmmx et Chihmis) published, under the name of Guatterm, a genus with exactly
the same oliaraoters as Aublet's Cananga. Unless, therefore. Hook. f. & Thomson aro
right in making a special case in establishing, as a gemis in the Linnoean sense, the
¿manga of Kumphius, Aublet's genus Cananga must stand, and to it must bo relegated
all the American specics referred to Rniz and Pavon's genus Gnatteria. Authorities vary
in their treatment of the Oananga of Rumphius. Danal (in his MomgrafUe de ta famille
dot Anonaeees) pronounces for the suppression of Aublet's 0,manga in favour of that of
Rumphius, who, he incorrectly says, assigned too species to i t ; the fact bemg, as already
shown, that Rumphius divided Cananga Into (a) cultivated (with one sort) and (Í) wdd
{tglcslres) with three sorts. Dunal {and I think wrongly) refers all tho Canangas of
Rumphius, to Umna. In their Genera Plaiilarmu, Mr. Bonthnm and Sir J. D. Hooker
retain the Cananga of Rumphius and reduce Cananga of Aublet to Quatttria. Bâillon, on
tho other hand, retains the Camnga of Aublet as a genus, and to it refers all the South
American spcoies of OuatUria. Ho reduces Cananga adórala, H. f. & Th., to Umna, and,
altering the termination of its gcnerio name, he makes it a section of Unona under the
sectional title of Cananrjiim.
The gi-ounds for separating Cananga from Umna as a genus are thus stated by the
authors of the Flora Indica : " In habit and general appearance this genus closely
resembles Umna; but the indefinite ovules prevent its being referred to that genus. The
peculiar stamen (with a long oonica! apical point) and tho seeds are themselves, wo think,
sulHoient to justify us in distinguishing it as a genus." The simplest solution of the
synonymic knot, and one for which there is some justification on the ground of structure,
appears to lie in the acceptance of BaiUon's suggested name and in giving up that
of the authors of tho Flora Indica.
The synonymy of Gnatteria is further complicated by the fact that a largo number
of species with valvate aiStivation were referred to it by Wallich and others. These,
however were sepai-ated by Hook fil. & Thomson, by whom tho genus Pohjaltlna was
fonned for their- reception. Sii- Joseph Hooker refers to Cananga, not only the species
C. Oihrata but another named C. virgata. The latter plant appears to me, m tho light
of full material recently received, to be a typical Cjalhocalgx, and to that genus I
AK0NACEJ3 0¥ EEITISH ISDIA. **
bave ventured to remove H. A third species, ^ J 1 1 s ^ pi - - r RJIN ^^cds both
no good ^P«?"™-) are peculiar. I quote tho following excel ent
of this species and o C. o d a ^ u , ^^^^
aescaptmn of tho o ^ p S like those of tl,o soction of and
page 1 3 0 : - The » j „j lirownish-yellovv, brittle testa is
t S r ^ r - W e : w S ' p e n e t r a t e i . o the albumen, taking the placo of
Z ilat plates which are found in the rest of the order.
Flower. 2 or 3 in. long
Flowers 1 to 135 in. long
1. C. odoratimi.
2. C. Sccrteohinii.
60 feet high ; young membranous, ovato-oblong or oblong-
W i'X; s r i r-:
X i -a ». -li- - "!"•
b:;™'-; r s =
,r f I Tl, Fl Imi 130' FI. Br. Ind. i. 56; Mq. I I . Ind. Bat. i. f t . 2, 40,
k J ' f Ì mZJI. 3-Uvaria odorata, Lané. III., t. 405, / . 1; Mo.L Fl. Ini.
r i ^ - F ^ F l ^ : Ln., t. 18; Grif. Mtul. iv. 7 1 2 - U . fracta, Wat,
I w 6 4 6 0 - U . axillaris, toS. Jnd.il 667.-Unona odorata and U. loptopctala, Dumi
Amn. 108 mul 114; DC. Frcir. i. 90 and 91; Delcss. le. Sol., t 88.
In all tho provinces, planted. Indigenous in Tenassenm, Java, and tho Philippines.
F j ™ 67 Canangium odoratum, Baillau 1, Flowering branch; 2, ripo carpeb;
3 & 4, sections of ripe c a r p e l s - » / natural si,o ; 5 & 6, sections of ovaries; 7, ripe s e e d -
onlarge'd. (Ne. 1 is copied from Slnme Fl. lav. Amn., t. 9.)
2 CMAKGiml ScoBTEonmi, Kiig in Jomn. 4s. Soc. Bengal 01, pt. 2, 42. A tree
30 to 40 feet high; young branches puberulous but speedily glabrous, dar-k-colonred and
lonticellate Lcaoes membranous, broadly'ovate, acute; the base broad, rounded, slightly
obliquo; both surfaces pubescent when very young, ultimately glabrescent, glandulard
o t t e d - ' t h e midrib and 6 or 7 paii-s of nerves adpressod-pubescent ; length 2-5 in.,
Asx. Uov. Bor. Gaud. C\Lcen-A Vol. IV.