A REVISION OF SALMEA 201 
in Salmea ; of solitary axillary and terminal heads, or these few 
2-6) on bracteolate pedicels on a common peduncle, in Spilanthes), 
are strengthened rather than weakened by the inclusion of 
S. pauciceps in Spilanthes. Deapite J acquin’ . onatnng mer e pbic 
. the stems of S. figs ache as ‘“fruticosi,” a character well show 
by specimens in the British Museum collected in Cuba by Ruge al 
(No. 6), der eat ao seem to have determined that all the 
species of the genus must be ——— bing pa are described 
by Bentham (Gen. ii. 381) as “herb a v. rarius perennes,’ 
by Hoffmann (Nat. Pfi. iv. pt. 5, 23 7) aa x « Kréuter,” pe, b 
Moore in his revision of the genus (Proc. Am. Acad. xlii. 522 
(1907) ) as ‘herbaces.”’ Aside from these two shrubby species, 
however, which c wie Bt a referred to Salmea without destroying 
the distinction betw sa two genera, all the species of 
Spilanthes seem to be pS : 
- petrochiotdes Hook. ‘& teak (sphalm.) = S. PETROBIOIDES 
Gri 
S. pteroboides Millsp. (sphalm.) = S. PETROBIOIDEs Griseb. 
S. salicifolia Brongn. = 8. scanpENs (L.) DC. 
II. Saumeopsis Benth. 
eads discoid, whitish. Involu ‘ge Atay ate, graduated ; 
phylaties eee ovate to (inner) oblong-ovate, indurate below, 
oriaceous-chartaceous, densely striglose -pilosulose, in age 
nt lirihatas with short narrow loose or spreading herbaceous tips. 
Receptacle convex. Anther-appendages oval, rounded at apex. 
pla: 
Achenes eilicluta. Pappus of 2 subequal awns and 8-14 slender 
acuminate squ uamelle, united about to middle, and at base to the 
awns. Otherwise as in Salmea.—Salmeopsis Benth. in B. & H. fil, 
Gen. Pl. ii. 381, no. 412 (1873); in Hook. Ic. xii. 47, t. 1152 (1873) ; 
Baker in Mart. Fl. Bras. vi. pt. 3, (1884). 
Craussent Benth. in Hook. Ic. xii. 47, t. 1152 (1873). 
cence. 
at ners shallowly Tet are 6-7 pairs, depressed, 
13-19, < on :2 Scheadod cyte or peacaite peduncles 06-85 cm 
mm. 
tities end at re ieess bs AzIL: Minas Gerace, 1840, Claussen 
(types: Kew). The specimen from Pert ay (Balansa, at in 
hb. ‘Kew) referred to this ies by Bixer i in the Flora Brasili- 
ensis proves to belo scandens (L.) DC., and the weed 
ng to 
is confined to Brazil so far as known at present. 
