512 MR. G. BENTHAM— REVISION OF THE GENUS CASSIA. 
hand, is an annual, with small mostly pentandrous flowers, and is abundantly diffused 
over tropical Africa, Asia, and Australia, but as yet unheard of in America, excepting in 
a single locality in Jamaica, where it was rare, probably introduced, and from whence it 
has not spread. These circumstances might suggest a separate divergence from the 
Chamecristas ; and indeed Schrank proposed a genus Grimaldia, consisting of C. Absus 
and two or three Chamecriste, with small flowers and a few leaflets. But the affinity 
of C. Absus certainly appears closer with the American C. hispidula. The American 
Absi generally run perhaps more directly into Chamecrista, through C. ( Absus) coriacea 
and C. (Chamecrista) Burchellii. 
The last section, Chamecrista, with a large number of herbaceous or suffrutescent 
species, readily propagating by seed, and many of, them very abundant in individuals, is 
an exceedingly puzzling one to botanists. "The nicest shades by which the majority of 
forms pass into each other make it impossible to settle what is to be regarded as species 
with any satisfaction; and although I have reduced a considerable number of Vogel's, 
and adopted a few others which he had suppressed, I have not yet much confidence in 
the result, and in many instances his judgment may have been better than mine, although 
I have had an infinitely greater number of specimens before me on which to base my 
conclusions. I have thus enumerated fifty-two American and sixteen Old World species, 
only one of the latter being apparently identically represented in America, whilst the 
only American species extending into the Old World appears to be the N. American 
C. nictitans, which, if the small specimens in the Hookerian herbarium are correctly 
determined, is found also in N. China and the Himalayas. 
Two of the most distinct groups, Xerocalyx and the Coriaceæ, are exclusively Ame- 
rican ; so also are the.greater number of the species with few leaflets; but the great mass 
of forms allied to C. Chamecrista in the New World and to C. mimosoides in the Old show 
divergences in different directions with different correlations of characters which render 
it impossible to distribute them into geographical groups. Thus, venation of the leaflets 
being one of the most constant characters in the section, we have the nearly central 
costa of C. Chamecrista marking several American species, and the very excentric or almost 
marginal costa of C. mimosoides in several of the Old-World species; but in the case of 
two species, which at first sight, as well as on examination of the flowers, fruits, and 
all but the leaflets, appear identical, the American C. patellaria has always the venation 
of the Asiatic and African C. mimosoides, whilst the ‘Asiatic and African C. nigricans has 
the venation of the American C. Chamecrista. 
Although I have seen nothing from the Old World corresponding to the widely spread 
American C. Chamecrista or C. glandulosa, two of the numerous forms of the Asiatic and 
African C. mimosoides appear to be represented in America: one is the C. hecatophylla 
figured by Colladon, from one of the W. Indian islands, and afterwards correctly reduced 
by De Candolle himself to C. mimosoides, of which I have seen a single specimen amongst 
Appun’s Guiana plants; the other is the C. Æschynomene, DC., more common in tropical 
America than C. hecatophylla, but still apparently not abundant, which I am unable to 
distinguish. from : tho Asiatic C. Wallichiana, aa admitted to be a — pd i 
C. mimosoides. 
