TEBN8TRCEMIACE2E. 



24 L 



In other species of the same genus, the petals are more or less 

 clearly alternate with the sepals ; in others the flowers are poly- 

 gamous dioecious ; l the style and its stigmatiferous divisions are 

 very variable as to form and dimensions f the number of the ovary 

 cells is three or four, and they are bi- or tri-ovulate ; the pericarp is 

 sometimes thin and almost membranous, and sometimes thick and 

 suberous. 3 But in all the known species, some twenty 4 in number, the 

 stem is arborescent or frutescent ; the leaves alternate, persistent, 

 entire or dentate, coriacous, exstipulate. The flowers are axillary, 

 pedunculate, solitary or disposed in cymes ; and their calyx is 

 accompanied by two or three bracts resembling the sepals, but 

 smaller. Three parts of the known species are natives of tropical 

 America ; the rest of the warmest parts of Asia and the Indian 

 Archipelago. 



Beside Ternstrcemia are placed the nearly allied 

 genera Adinandra, Eroteum, and Etirya, only sepa- 

 rated from them in an entirely artificial manner : 

 the first, because its seeds are small and numerous 

 instead of being large and few in number ; the 

 second, because in its small flowers, often poly- 

 gamous or dioecious, with petals free or scarcely 

 united at the base, the ovules, indefinite in number, 

 are inserted towards the middle of the internal angle 

 of the ovary cells ; the third, because its dioecious 

 flowers have generally an oligandrous androceum. 



In the two genera Visnea and Anneslea, the general organization 

 is the same ; but we make of them a small subseries (Visneea:), 

 because their floral receptacle, instead of being convex, becomes 

 more or less concave ; the insertion of the perianth and the andro- 



Visnea Mocanera. 



Fig. 264. 

 Induviate fruit (j). 



1 This is found especially in T. penangiana 

 Chois., which has been made the type of the 

 genus Erythrochiton (Griff., Notul., iv. 565 ; 

 — Chois., in Mem. Gen., xiv. 126, nee Mart.). 



2 The divisions are very large in Erythro- 

 chiton, and radiating in Reinioardtia. 



3 It is divided into six cavities in Vcelckeria 

 (Kl. & Karst., ex Endl., Gen., Suppl., iv. 66; 

 — Chois., loc. cit., 125) ; but the ovary being 

 three-celled, they are supposed here to be only 

 half-cells, doubtless separated by false parti- 

 tions. 



* Sw., Fl. Ind. Occ, ii. 929.— Ruiz & 



VOL. IV. 



Pav., Prodr., t. 21. — H. B. K., Nov. Gen. et 

 Spec, v. 207, t. 463.— A. S. H., Fl. Bras. 

 Mer., i. 231. — Moric, PI. Nouv. Amer., t. 

 12, 13.— A. Rich., Fl. Cub., t. 27-— Wight, 

 Icon., t. 47 (Cleyera). — Sieb. & Zucc, Fl. 

 Jap., t. 80. — Miq., Fl. Ind.-Bal., i. p. ii. 

 470. — Griseb., Fl. Brit. W.-lnd., 103; Cat. 

 PL Cub., 35. — Turcz., in Bull. Mcsc. (1858), 

 i. 211; (1863), i. 577.— Seem., Voy. Her., 

 Bot., 87. — A. Gray, Amer. Expl. Exp., Bot., i. 

 209. — Tr. & Pl., in Ann. Sc. Nat., ser. 4, 

 xviii. 258. — Walp., Sep., i. 368; ii. 804 ; v. 

 130 ; Ann., iv. 341 ; vii. 361. 



It 



