538 MR. G. BENTHAM—REVISION OF THE GENUS CASSIA. 
Seetio 5 . CHAMASENNA. 
Frutices, rarius arbores, v. in Brachycarpis Pictisque paucis herbs. Folia inter- 
glandulosa v. eglandulosa, rarius nulla (ad squamas minutas reducta). Stamina perfecta 
7 (rarius 6 infimo minore imperfecto), staminodia 3. Legumen compressum, ad suturam 
utramque dehiscens; valvulæ nunc plan: marginibus incrassatis v. nerviformibus, nune 
convex: vix marginatæ, interdum medio ad semina elevate subcristate v. longitudinaliter 
alate. Semina transversa v. obliqua, valvulis parallele compressa, rarius oblongo-sub- 
teretia v. leviter septis parallele compressa. 
Like Chamefistula, Chamesenna as a whole is an artificial section, which, as a matter of convenience, I 
have adopted from De Candolle and Vogel with some slight modifications. "The following groups as 
between themselves appear to be more natural, although not always definitely characterized ; but the 
Platycarpe, for instance, pass into the Chamefistule Corymbose, and are often undistinguishable from them 
except by the pod, and the Floride Paniculate are represented in Chamefistula by the Excelse. There are, 
however, no Chamesennas corresponding to the Bacillares ; for, in the few cases where the leaflets are 
reduced to two pairs with a gland between the lowest as in C. multiflora or C. glaucescens, their shape and 
general habit are totally different from those of the Bacillares ; and in Chamesenna neither the Leptocarpe, 
Picte, or Brachycarpe, nor the Floride (except the Paniculate) are at all likely to be mistaken for 
Chamefistulas. 
A. Folia interglandulosa (glandula inter par infimum, rarius inter paria plurima v. 
omnia foliolorum), rarius nulla. 
Series 1. Pachycarpæ. Anthere erostres obtuse v. poro (simplici v. duplici) parum pro- 
minulo v. in cupulam sessilem expanso. Legumen nunc breve latumque, nunc longius et 
angustius, valvulis tunc. sepius convexiusculis marginibus latis v. parum prominulis, in 
perpaucis ei Rostratarum simile. Foliola pauci- v. rarius multijuga, obtusa v. acutius- 
cula, sepe crassiuscula. Stipule (excepta C. stipulari) lineares setacec v. caducissime. 
The species of this group are none of them Brazilian or Guianese; and some of them are extratropical. 
They extend along the western side of the Andine range, without ascending high into the mountains, 
from temperate Chili to Peru, reappearing in Central America and Mexico; and one species, C. baha- 
mensis, which connects the group with the Leptocarpe, is found in two distinct varieties, or perhaps 
species, in Florida and in the Bahamas. The group generally shows a pod less flattened than that 
of other Chamesennas, and often nearly that of Oncolobium; in some species, however, it is very flat 
but broad and short, with thin edges; very rarely, as in C. bahamensis, it seems to vary in the same 
species from the narrow flat pod with raised margins of C. : biflora to the shape more common in Pachy- 
carpe, with convex valves not bordered. Although, therefore, I have named the group after the pod, I 
have taken the most definite distinctive character from the anthers; there are a few species, indeed, in 
which the pod is unknown, and which are therefore only conjecturally placed in the group. 
* Foliola acuta, submembranacea. Species Chilenses. 
90. C. STIPULACEA, Ait. ! Hort. Kew. ed. 1. ii. 52. Foliola 4-8-juga, ovato- v. oblongo- 
lanceolata, acuta, glabra v. pubescentia. Stipulæ foliaceæ. Legumen breve, latum, tenue. 
C. fetida, Ruiz et Pav. !, G. Don, Gen. Syst. ii. 443. 
C. opaca, Grah. ex char. in Vog. Syn. Cass. 42. 
Hab. Extratropical South America: Chile, near Valparaiso. 
The broad stipules are those of the 4uriculate ; but the pod and the habit are quite different. I have 
been unable to find Graham’s description of C. opaca in the "Mdisinagh;Nos:Fbilosiphiesl Journal. 
