ANONACE^. 



197 



concave, caducous petals. Tlie stamens and multiovulate carpels are 

 exactly those of Uvaria. The fruit is as }'et unknown. Only one 

 species of this genus has as yet been described,' a shrub from Borneo, 

 whose flexible twining branches are covered with alternate leaves, and 

 whose flowers are arranged in dense leaf-opposed or lateral spikes. 



In Canau^a,- the flowers are constructed on the same plan as in 

 Uvaria, differing only in a few characters of secondary importance. 

 The receptacle is convex,"* bearing 

 in succession a calyx of three sepals, 

 valvate in the bud, and two corollas 

 each of three equal, or nearly equal, 

 petals imbricated in the bud, and 

 spreading on the expansion of the 

 flower. The stamens are very nu- 

 merous, and closely packed in a 

 spiral; their anthers are extrorse, 

 with two parallel cells, surmounted 

 by a truncated dilatation of the con- 

 nective. Above the androceum, the 

 apex of the receptacle presents a 

 circular platform, often surrounded 

 by a slightly projecting rim. On this 

 surface are inserted the indefinite car- 

 pels, each consisting of a one-celled 

 ovary, surmounted by a nearly sessile swelling covered with stigmatic 

 papillae. At the base of the ovary is a placenta, bearing an as- 

 cending or erect ovule, with its micropyle outwards and downwards. 

 The fruit is multiple (fig. 23.'2), consisting of an indefinite number 



Cananga {Giiatteria) Schomhurgkiana. 



Fig. 232. 



Fruit. 



1 T. volubile MiQ., loc. cit. Nearly all the 

 organs of this plant are covered with rufous hairs, 

 like those of most Ucarias. 



2 AUBL., Guiaiu, i. (1775) ml, t. 244 

 (Rttmpu., Herb. Amhohi., ii. (1711) 195, t. 05; 

 Hook. F. k Thoms., Fl. Imh, i. 12'J).— Gmcti!- 

 teria R. & Paa'., Prodr. (179 1) 85, 1. 17 (in Auctt. 

 Flor.Asiaf!c.).—l>vy., Mon., 50, t. 30-32.— DC, 

 Sgst., i. 502; Prodr., i. 93. — E>VUL., Gen., n. 

 4721. — B. H., Gen., 23, n. 7. We have explained 

 (Adansonla, viii. 336,) why the gei'.uric name of 

 AUBLET should in any cjtse retain its priority, 

 though it does not apply to the same plants as the 

 genus Cananga of Rimpur's. These last are, in 



our opinion, Unonas, and th(! name of Guatteria 

 was only created by Ruiz & Pavon nineteen 

 years after the publication of Aublet's work; 

 so that the name Cananga, as applied by 

 ArBLET, has still nineteen years' priority. 



"* It is often dome-shaped, except in that su- 

 perior truncated part [torus apice tnincatus), 

 forming a sort of platform which bears the 

 gyna?ceum, and is surrounded by a small annular 

 pad, of which we shall speak below. This must 

 be considered nothing else than the first rudiment 

 of the receptacular sac, found so greatly de- 

 veloped in most species of Xylov'a. 



