CASSIA. 
229 
in the 2 preceding- sp., are flattened vertically or parallel to the 
transverse partitions, not as usually in Leguminosce laterally or 
parallel to the valves of the pod. 
§ II. Chamcesenna DC. Shrubby. Pod compressed dehiscent 
lengthways, not pulpy within. Seeds horizontal. 
fft4. C. tomentosa Lam. 
F ulvous-tomentose, the branches velvety ferruginous ; lfts. in 
4-8 pairs oval oblong more or less obtuse mucronulate densely 
tomentose beneath, shining and smoothish above, their edges 
thickened and subreflexed, with a gland mostty between each 
pair \ rac. axillary short corymbose 3-6- or 7-fl.; pod compressed 
linear-oblong velvety-pubescent. — DC. ii. 496 ; Vogel Syn. 22. 
C. multiglandulosa Jacq. Ic. Par. i. t. 72. — Shr. per. Mad. cult, 
reg. 1, 2, r. Here and there in gardens or about cottages. At the 
Mount, and at Prazeres, &c., to the westward. Throughout the 
year. — A bushy shr. 4-6 ft. high, with a distinct st. and hand- 
some foliage. Branches flexuose strongly grooved or ribbed and 
angular, the younger densely clothed with fulvous ferruginous 
velvety pubescence like the petioles. Lfts. very dark full shining 
gr. and minutely downy above, pale and densely ferruginous- 
tomentose beneath. No gland at base of petiole. FI. very hand- 
some, rich golden- or or.-y., in stalked axillary close bunches 
shorter than the 1. Pod 3-4 in. long, 4-5 lines broad, straight 
abruptly beaked or mucronate much compressed or flattened, 
the sutures slightly margined with a raised hem-like border on 
each side, quite dry or pulpless within. Seeds numerous hori- 
zontal i.e. flattened vertically, tawny or olive-brown. 
C. occidentalis L., C. sulfurca DC., C. australis Sims, and C. 
glandulosa L. or C. Chamcccrista L. occur sometimes in gardens 
about Funchal ; and I have myself occasionally raised the first- 
named sp. from W. I. (Jamaican) seeds. But none of these 
have ever showed any tendency to become diffused or natu- 
ralized. 
C. ruscifolia Jacq. Ic. Rar. i. t. 71. (DC. ii. 498-), given by its 
original describer as a Mad. pi., is probably not distinct from C. 
occidentalis L. In foliage it somewhat resembles C. Icevigata or 
C. Jloribunda ; but it has a wholly different habit, and the 
petiole is furnished with a gland at its base. The occurrence of 
Jacquin’s pi. in Mad. can have been in any case merely tempo- 
rary or accidental. 
The Judas-tree, Cerois Siliquastrum L., remarkable for its simple 
orbicular-heartshaped 1., by which De Candolle considers it 
allied to Bauhinia, though placed by Endlicher at the end of 
Sophorea, is a conspicuous and frequent ornament to gardens 
M 2 
