Saxifragai i - ! 



r.Ki.M \< i i 





Obder CCXVI] BREX1 .CI I I 



it,. 



Diagnosis. < • / 



Trees, with nearly simple trun I •.,■, coriac s, al 



with deciduous minute Btipules. Flowers green, in axillary umbel 



.small, persistent, 

 cated. Petals 5, I 

 vation. Stamens '.. In; 

 with the petals, arising from a 

 which is toothed betw 

 there oval, innate, 2-celled, but 

 dinally, fleshy at the apex ; pollen ti 

 cohering by means of fine tin 

 superior, 5 celli I, with nun. 

 attached in two rows t" placentas in die 

 Btyle I, continuous ; stigma aim| 

 drupaceous, 5-cornered, "marked with nu- 

 merous small Bcarcel) elevated papill 

 surface of an oran| lied, manj 



Seeds indefinite, horizontal, smooth .. . 

 ing, brown, ovate, slightly angular, about die 

 size of those ol a raisin, attached t.. the 

 with a double integument, the inner •■: which 

 is membranous; cotyledoi 

 radicle cylindrical, centripetal; (albumen 

 Hi shy, according to Thouai 



There exists in Ma.la_-a-.-ar a 

 Brexia, of which tin- above i- .. 

 taken in part from the living plant, and i 

 from Dr. Wallich, in th 

 The position which this plant ought I 

 Indeed, we are not certain wh 

 Beeds have albumen ; for although Thouare states it to be fleshy, Dr. Wal 

 upon the subject, and no other Botanist seems to have examined the 

 is that of some Ardisiads, especially of Theophrasta, from which it differa in 

 polypetalous, in the stamens being alternate with the petals, and in main 

 Btances. \\ ith Rhamnads and Spindle-trees its relation is no doubt - 

 are indefinite. Some rest mblance may I"- traced between it and Ai 



'■i ill\ m the resinous appearances visible upon tin- young s| ts,and al 



fructification is entirely at variance with that Order. With Pitt 



its hypogynous definite stamens, its polyspermous fruit, its alternate und 



and habit ; hut it is probable that thr embryo i- not such as b fits I 



ncher places it at the end of Saxifrages, combining with it lv i 



the latter a genua having the ovary adherent in Bome d< 



approximation is right, and it certainly Beema probable, and it" ti 



Bhould prove, when reexamined, to have albumen, as Thouare 



reticulated testa represented by Gsertner, then Brexia must 



form, and map be Jooked upon as a genus of the Saxifragai Uluu 



h. -t tn place it tin- the present Nor can I doubt that H 



bis / onographia, is of the same class, notwithstanding il 



i- quite analogous to the toothed disk of Brexia. 



QBN1 R \ 

 herba, .1. Curm. im, Lam. 



Brexia, Thouart. \ iphyllum, / 



Numbers. Gen. I. S 



I < i ' I WW 111. 



in a natural classification is unsettled. 



Position. 



. ■ 



' ■ < i,\ \ \\ ill i Brexia ma I 



