192 ERICACEH (Guthrie & Bolus). [ Hrica. 
incurved, nearly 2 lin. long, erested; pore 1-2 the length of the 
cell; crests narrow-lanceolate, acute, pendulous, long-ciliate, brown, 
about § the length of the cells; style ineluded, 4-gonous below, 
swollen and terete above; stigma subsimple, small ; ovary glabrous, 
or thinly hispid on the apex. 
Sourn AFrica: without locality, Masson! Niven ! 
Coast Reeion: Caledon Div.; rocky summit of the Genadendal Mountain, 
4800 ft., Bolus, 5421! Schlechter, 9835 ! 
This species does not come very happily into this or any section. Though 
distinct, it is allied to E. lachneoides (E. Lachnea, var. purpurea, Andr., as 
figured in Andr. Heathery, t. 170). Scarcely any of the specimens, excepting 
Schlechter’s, exhibit fully-developed flowers, 
268. E. virginalis (Klotzsch ex Benth. in DC. Prodr. vii. 653) ; 
apparently a small slender plant, with filiform, white-tomentulose 
branches (“quite glabrous,” Bentham); leaves opposite, erect, 
adpressed, oblong, obtuse, keeled, glabrous, glossy, under 1 lin. long; 
flowers 2-nate, corolline ; pedicels decurved, nearly 1 lin. long; 
bracts subremote, small ; sepals narrow-lanceolate, acute, adpressed, 
round-backed and sulcate, or “keeled,” thickish, under 1 lin. long ; 
corolla subtubular, 4-gonous, “red,” 2 lin. long; limb erect or 
“ stellate-spreading 1” (Bentham); segments rounded, imbricate at . 
the base, 1-1 the length of the tube; filaments slender, tapering to 
the apex, about 12 lin. long ; anthers narrow-oblong, obtuse, some- 
what rough, under 2 lin. long, crested; pore * the length of the 
cell; crests ovate, bluntish, about 4 the length of the cell ; style 
apparently included ; ovary sessile, glabrous, 
Coast Reeron: Tulbagh Div. ; mountains near Tulbagh, Ecklon ¥ Zeyher! 
We have seen and examined a frustule with a flower from a specimen marked 
as above, which is very probably the type. We find, however, no trace of a 
stellate-patent corolla-limb, a character which is often deceptive and lost in dried 
specimens, and which Bentham mentions with a sign of doubt. In other 
respects it does not agree well with the § Zamprotis where Bentham provisionally | 
placed it, the flowers being distinctly corolline ; but it fits better and fairly well 
here. The species is a very distinct one. 
Section XX. PACHYSA. (Sp. 269-292.) 
269. E. ramentacea (Linn. Mant. 65); erect, 1-1} ft. high; 
mostly much branched from shortly above the base; branches usually 
much spreading, sometimes erect, puberulous or glabrous; leaves 4- 
(rarely 3-) nate, erect or spreading, linear, very slender, glabrous, 
15-3 lin, long; flowers terminal, umbellate, with a heavy, somewhat 
musk-like odour; pedicels erect or spreading, slender, red, 13-1} lin. 
long ; bracts 2 subapproximate, 1 median, linear, red ; sepals lanceo- 
late, keeled, glabrous, naked or ciliate, reddish, about 2 lin. long; 
corolla mostly subglobose, occasionally ovoid, contracted at the 
throat, glabrous, slightly viscid, crimson, 1} to nearly 2 lin. long ; 
limb short, erect, darker-coloured ; anthers included, dorsifixed near 
