110 SELAGINE® (Rolfe). [Hebenstreitia. 
Drege, Zwei Pflanzengeogr. Documente, 76, 212; Walp. Rep. iv. 144; 
Choisy in DC. Prod. xii. 3. P. lanceolata, var. B pubescenti- 
hirsuta, Walp. Rep. iv. 144, 
Coast Recion: Piquetberg Div.; in stony places on the Piquetberg Range, 
1500-3000 ft., Drége ! 
bs Polycenia lanceolata var. B. glabrata (E. Meyer, Comm. 245, and in Drége, 
Zwei Pflanzengeogr. Documente, 125, 160, 212; Choisy in DC. Prod. xii. 3. 
2 P. lanceolata var. glabra, Walp. Rep. iv. 144), is based on material collected by 
Drége in Coast Reeton: Knysna Div., Groene Valley, 500 ft., and Eastern 
Reeion: Natul, below 100 ft., which I have not seen. It is described as having 
more or less glabrous branches and leaves, very glabrous bracts, and the fruit 
as having only one spurious cell in each achene. 
* 
> 
+ 
— boa: it: : 
a og 
er 4 
_ 29. H. leucostachys (Schlechter in Engl. Jahrb. xxvii. 186) ; 
_ perennial, somewhat branched, 1-1} ft. high; branches erect or 
ascending, pubescent; leaves oblong-lanceolate, acute, acutely 
_ dentate, more or less pubescent, 3-1; in. long; spikes elongate, 
3 in. or more long, mauy-flowered; bracts oblong-lanceolate, acute, 
somewhat pubescent, 23 lin. long; calyx oblong, with 3 acute teeth, 
3-nerved, 2 lin. long; corolla 5 lin. long: tube slender; lobes 
oblong, inner rather narrower and longer than the outer; fruit 
oblong, obtuse, achenes equal, semiterete, glabrous (Schlechter). 
Coast ReGion: Worcester Div.; Bains Kloof, 1300 ft., Schlechter, 9158! 
I have not seen the fruit, but the flowering specimen is so similar to the 
preceding species that I think it must belong to this section. It chiefly differs 
from H. lanceolata in its looser pubescence, less hairy bracts, and more strongly 
toothed calyx. Schlechter compares it with H. dentata, Linn., which belongs to 
Eu-Hebenstreitia, but as he has labelled a good fruiting example of H. Dregei; 
x. Rolfe (Polycenia fruticosa, E. Meyer), “H. crassifolia, Choisy, var.?”” (which 
also belongs to the other section) it would appear that he did not examine the 
structure of the fruit. 
30. H. cordata (Linn. Syst. ed. 13, ii. 420); perennial, much 
_ branched, 1 ft. or more high; branches hispidulous; leaves 
_. numerous, spreading, broadly ovate or cordate-ovate, obtuse or sub- 
_ mucronate, somewhat fleshy, 1-3 lin. long; spikes oblong or some- 
what elongate, 1-2} in. long, very dense; bracts ovate, mucronate, 
23-3 lin. long ; calyx oblong or obovate-oblong, obscurely 3-denticulate, 
27-8 lin. long ; corolla white with yellow throat (Burchell), 4-5 lin, 
long; tube stoutish ; lobes broadly oblong, subequal; fruit ovoid- 
‘ globose, 2-2} lin. long. Linn. Mant. 420; Murr. Syst. Veg. ed. 14; 
* 570; Lam. Encyel. iii. 78; Ait. Hort. Kew., ed. 1, ii. 356, ed. 2; 
iv. 48 ; Thunh, Prodr. 103, and Fl. Cap. ed. Schult. 478; Willd. 
Sp. Pl. iii. 332 ; Jarosz, Pl. Nov. Cap. 13; Choisy Mém. Selag. 23; 
Spreng. Syst. Veg. ii. 754; Rolfe in Journ, Linn. Soc. xx. 347, 356. 
Polycenia cordata, E. Meyer, Comm. 245, and in Drege, Zwei 
Pflanzengeogr. Documente, 94, 129, 212; Walp. Rep. iv. 143; 
Choisy in DC. Prod, xii, 3. 
Soutn AFrica: without precise locality, Bowker ! Wallich / 
é _ Coast Recion: Malmesbury Div. ; Hopefield, Bachmann, 1151! Cape Div. ; 
on the sea coast at False Bay, Bolus, 2890! about Camps Bay, Burchell, 3051 
sea shore at Camps Bay, Harvey, 513! Cape Peninsula, Railway at St. James, 
s 
oe. 
