182 ERICACER (Guthrie & Bolus). [ Hrica. 
about 1 the cell in length; style included, short, thickened towards 
the apex ; stigma clavate-capitate, large; ovary glabrous. 
Coast Rearon: Worcester Diy. ; on Sneeuwkop Mountain, near Bains Kloof, 
Marloth in Herb. Bolus, 6497! 
A very well-marked species, if its characters are constant ; we have only one 
small specimen from the single station cited. It verges towards the Subgenus 
Platystoma, and is distinguished in the present section by the combination 
of a corolla-limb rather large for this section, with a very small cristate 
anther. 
250. E. quadrangularis (Salisb. Prodr. 297, not of Andrews) ; 
erect, 12-18 in. high; branches mostly much spreading or divari- 
cate, rarely subvirgate, puberulous; leaves usually spreading, or 
only the younger erect, linear, glabrous, the younger ciliate, $-1} lin. 
long; flowers 4-nate; pedicels slender, about 1 lin. long; bracts 
remote, minute; sepals ovate or lanceolate-ovate, acute, subsearious, 
ciliate or naked, keeled, glabrous, }—% lin. long; corolla generally 
broad-cyathiform, more rarely subcampanulate-cyathiform, or the 
tube sometimes obeonic (“ poculiform,” Salisbury), mostly widened 
at the mouth, sometimes tetragonous, 1-14 lin. long, white to rosy ; 
limb erect, variable in shape and length, 1—3 the length of the tube, 
the interstices of the limb-segments acute at the base; anthers 
included, from obliquely cuneate-oblong, 3 lin. long, in Salisbury’s 
type to subtriangular with very rounded angles, } lin. long and 
proportionately broader (the front view of the anthers showing 4 
greater width than their length), in most other specimens quite 
smooth and glabrous, aristate ; pore 3—2 the length of the cell; awns 
narrow, rough-edged, but not fringed, about 3 the length of the cell ; 
style straight, included; stigma capitellate, mostly small; ovary 
glabrous. EH. quadreflora, Salisb. in Trans. Linn. Soc. vi. 375; 
ane 3 DC. Prodr. vii. 679. E. corusca, Lichtenst. ex Benth. 
.c. 679. 
Sout Arrica: without locality, Rorburgh! Herb. Salisbury ! 
Coast ReGion, between 150 and 1500 ft. : Clanwilliam Div.; near Clan- 
william, Leipoldt, 211! Worcester Div.; Matroos Berg, Marloth / Paarl Div.! 
French Hoek, Bolus, 6983! Stellenbosch Div.; Lowrys Pass, Schlechter, 5353 5 
Caledon Div. ; near Zoetemelks Vlei, Guthrie, 3285! near Caledon, Bolus, 
6749 ! Schlechter, 5604! Bredasdorp Div.; near Elim, Bolus, 6734! Swellendam 
Div. ; Zuurbraak, Galpin, 3673! Riversdale Div.; near Riversdale, Schlechter, 
1932! Knysna Div. ; Millwood, Tyson! 
Salisbury says “the flowers are inconspicuous, but in the whole genus very 
distinct,” neither of which statements are intelligible to us, although we have 
seen and dissected his type, from which our specimens differ only slightly in the 
anthers. If we are correct in referring them here, of which we have little doubt, 
the inflorescence is abundant and conspicuous, though the flowers themselves are 
not large, and the corolla greatly resembles that of several other species. Salis- — 
bury doubtless described from a single specimen, and possibly a poor one. The 
Index Kewensis, i. 870, puts this species down as a synonym of E. denticulata. 
251. E. cyathiformis (Salisb, in Trans. Linn. Soc. vi. 376); 
erect, much branched, 1-1} ft, high; branches puberulous, the 
