FLORA OF SOUTH AFRICA. 59 
Gnidium (among which, indeed, it has been sent to us) than 
like other Phylicas. Ecklon’s diagnose is very good. The 
leaves are generally opposite, sometimes, however, verticillate, 
or, especially on the young (tomentose) branchlets, alternate, 3 
lines long, with a very short, but distinct, pubescent petiole, 
in form and size resembling those of Erica carnea, only they 
are a little thicker. 'The flowers, coloured like those of 
Soulangia rubra, are 4 lines long, perfectly sessile, and col- 
lected in considerable number into terminal heads or clusters, 
of the size of a cherry. They are surrounded at the base by 
an irregular involucre, composed of ordinary, but smaller, 
leaves. "The tomentum, with which the whole outside of the 
calyx is covered, is white, woolly rather than silky (as termed 
by Ecklon,) almost without lustre; tube of the calyx cylin- 
drical, slightly funnel-shaped ; limb erect, divided into five 
narrow scarcely acute lobes, of one line in length, smooth 
and reddish inside; petals minute, squamiform, cucullate, 
covering the anthers, dark-coloured (brown or purple?) ; style 
equal to the calycine lobes, stigma obtuse. 
BRUNIACES. 
Brunia microphylla, Thunb. Fl. Cap. p. 207, DC. prodr. 2, 
p- 44, n. 15.—Inter rupes in summitate montium Baviaansk- 
loof, alt. 3000 (IV. B, b.) Dec. 1838. Krauss n. 778. 
Our plant (which was also sent us as a Thymelea) might 
almost as well be taken for B. phylicoides, Thunb., which | 
differs only in villose leaves, and capitula of the size. ofa 
pea. The young leaves are slightly, but distinctly, villous 
on the back, and strongly ciliated on the margin. Perhaps, 
therefore, the two species ought to be united. We must 
further observe, that they belong to the genus Brunia, not to 
Raspalia, to which they are referred by Brongniart and 
Ecklon (enum. p. 100) the ovary being, according to our 
reiterated examination, adherent to the calyx, with its infe- 
rior portion.— Raspalia teres, E. Mey. in Plant. Drég. is very 
like our plant, and perhaps not distinct. dap — 
