Eriocephalus.] COMPOSIT (Harv.) 201 
2. E. septulifer (DC.! 1. c. 145); leaves alternate, tufted, silky-sil- 
very, thickish, obtuse, shortly linear, or very rarely trifid, channelled ; 
heads umbellate at the ends of short, lateral twigs, the pedicels longer 
than the leaves ; “the inner invol. having very short projecting ridges 
(septa) within.” DC. 
Has. Cape Flats, Drege! W.H. H. (Herb. D., Sd., Hk.) 
Leaves 3-5 lines long, } line thick, silvery. Wool of the invol. coarse and fulvous, 
3. E. sericeus (Gaudich.) ; leaves often opposite, linear, entire or 
trifid, silky-villous ; heads sub-corymhose at the ends of the branches, 
inner invol. not septiferous within. DC.l,c. 145. EH. Africanus, Burm. 
Has. Cape: not uncommon round Capetown, according to De Candolle. 
Of this I know nothing, unless a specimen in Hb. Hook. without flowers and 
probably from a garden, be referable to it, 
4, E. capitellatus (DC.! 1. c. 146); leaves alternate, tufted, linear, 
entire and trifid, silky-silvery, obtuse, sub-terete; heads at the ends of 
short, lateral twigs, sub-sessile, small ; rays 1-3, short. 
Has. Zwarteberg and Zwart Ruggens, Drege! Hexrivier, Burke/ Zey./ 860. 
(Herb, Hk., D., Sd.) 
A slender, much-branched bush ; twigs filiform, closely covered with leaf-tufts. 
Leaves 2~4 lines long, 3 line thick, several in almost every tuft trifid. Heads 1 line 
long, on pedicels 1-2 lines long or less. Wool pale fulvous. 
5. E.? petrophiloides (DC.! 1. c. 146); stem and leaves quite gla- 
brous; leaves alternate, trifid above the middle, the medial lobe linear, 
entire, the lateral lobes mostly bifid, all callous-mucronate ; pedicels 
axillary, solitary, filiform, longer than the leaves; outer inv. of 6-8 
oblong, glabrous, basally connate scales, Hriocephalus, 446, Hb. Eckl. 
Has. In S. Africa (no station given), Drege/ Eckl.! (Herb. Sond.) 
A most remarkable species, unlike any other and possibly not of this genus ; but 
the fl.-heads, in the only specimens I have seen, have their contents eaten by 
insects, leaving merely the outer invol. and a dense tuft of discoloured wool. What 
may be the origin of this wool, whether from an inner invol. or from the achenes, 
remains undetermined. DC.’s specimens seem to have been equally imperfect, 
Leaves 1 inch or more long, the lobes 4~5 lines long, 4 line wide, all tipped with a 
white, hard point. Pedicels 1} inch long. 
6. E. punctulatus (DC. 1. c. 146); leaves opposite and alternate, 
scattered or crowded, linear-semiterete, acute or obtuse (occasionally 
unidentate at each side), thickish, glabrous, more or less impress-dotted; 
pedicels in terminal umbels and axillary, short or long ; outer invol. 
seales glabrous, widely membrane-edged, inner not septiferous; rays 
obovate, 3-toothed. : 
Van. a, pedicellaris; pedicels uncial ; leaves scattered. LE. pedicellaris, DC. 1. c. 
Var. 8, punctulatus ; pedicels semi-uncial; leaves crowded or scattered. £. punc- 
tulatus, DO. lc. zs 
Var. 7, tenuifolius; pedicels mostly shorter than the leaves ; leaves crowded. 
E. tenuifolius, DC. 1. ¢. 
Var. 5, filifolius (DC.); leaves linear-filiform, elongate, remote; pedicels 2-3 
times longer than the heads. £. tenuifolius; 8, filifolius, DC. l.c. 
Has. a, Little Namaqualand, Drege! 8, Slopes of the Kamiesberg, near Kaspars- 
