Saxifragales.] 



BREXIACE^. 



573 



Order CCXVII. BREXIACE^.— Brexiads. 



Brexiaceae, Ed. Prior, No. 95. (1830) ; Arnott in Edinb. Encycl. 104. (1832) ; Martins Cotispectut 

 No. 297. (1835) ; Endl. Gen. p. 823.— Roussseaceae, DC. Prodr. 7. 521 ; Endl. Gen. p. 823. ' 



Diagnosis. — Saxifragal Exogens, with consolidated styles, a many-leaved calyx, alternate 

 leaves, and no albumen. 



Trees, A^-itli nearly simple trunks. Leaves coriaceous, alternate, simple, not dotted, 

 with deciduous minute stipules. Flowers green, in axillary lunbels. Calyx inferior, 



small, persistent, 5-parted ; aestivation imbri- 

 cated. Petals 5, hypogynous, twisted in aesti- 

 vation. Stamens 5, hypogynous, alternate 

 mth the petals, arising from a narrow cup, 

 which is toothed between each stamen ; an- 

 thers oval, innate, 2-celled, bursting longitu- 

 dinally, fleshy at the apex ; pollen triangular, 

 cohering by means of fine tlireads. Ovary 

 superior, 5-celled, with numerous ov^uleg 

 attached in two rows to placentae in the axis ; 

 style 1, continuous ; stigma simple. Fruit 

 drupaceous, 5-cornered, "marked with nu- 

 merous small scarcely elevated papillae hke the 

 surface of an orange," 5-celled, many-seeded. 

 Seeds indefinite, horizontal, smooth and shin- 

 ing, bro\vn, ovate, shghtly angular, about the 

 size of those of a raisin, attached to the axis, 

 ^vith a double integmnent, the inner of which 

 is membranous ; cotyledons ovate, obtuse ; 

 radicle cylindrical, centripetal; (albmnen 

 fleshy, according to Thouars). 



There exists in Madagascar a genus called 

 Brexia, of which the above is a description, 

 taken in part from the hving plant, and in part 

 from Dr. WalUch, in the Flora Indica, 2. 314. 

 The position wliich this plant ought to occupy 

 Indeed, we are not certain whether or not the 

 seeds have albumen ; for although Thouars states it to be fleshy, Dr. Wallich is silent 

 upon the subject, and no other Botanist seems to have examined the seeds. Its habit 

 is that of some Ardisiads, especially of Theophrasta, from which it differs m being 

 polypetalous, in the stamens being alternate with the petals, and in many other circum- 

 stances. With Rhamnads and Spindle-trees its relation is no doubt strong, but its seeds 

 are indefinite. Some resemblance may be traced between it and Anacards, espe- 

 cially in the resinous appearances visible upon the young shoots, and also m habit ; but its 

 fructification is entii-ely at variance mth that Order. With Pittosporads it agrees in 

 its hypogynous definite stamens, its polyspermous fruit, its alternate undi\aded leaves, 

 and habit ; but it is probable that the embryo is not such as befits that Order. End- 

 licher places it at the end of Saxifrages, combuimg with it Ixerba and Argophyllura, 

 the latter a genus having the ovary adherent in some degree to the calyx. If this 

 approximation is right, — and it certainly seems probable, — and if the seeds of Brexia 

 should prove, when re-examined, to have albumen, as Thouars says, and the peculiar 

 reticulated testa represented by Gaertner, then Brexia must be indeed a perigynous 

 form, and may be looked upon as a genus of the Saxifragal Alhance, where it is perhaps 

 best to place it for the present. Nor can I doubt that Roussaea, figured by EndUcher in 

 liis Iconographia, is of the same class, notwithstanding its opposite leaves. Its great disk 

 is quite analogous to the toothed disk of Brexia. 



GENERA. 

 Ixerha, A. Cunn. I Venana, Lam. \ Houssea, Smith. I i?o»woa,Rbm.etSchlt. 



Brexia, Thouars. \ Argophyllum, Forst. \ Rousscea, DC. | Rousseauvia, Boj. 



Numbers. Gen, 4. Sp. 6. 



Fig CCCLXXXVIII. 

 in a natural classification is unsettled. 



POSITION.- 



CelastracecB. 



-Brexiace^. — Cunoniaceae. 

 MyrsinacecB. 



Fig. CCCLXXXVIII.— 1. Brexia madagaacariensis ; 2. section of its ovary ; 3. seed of Argophyllum. 



