29 
vel subsessiles, axi demum ad 5 (raro 8) mm. longo; bracteae 
ovatae, acutae, minutae. Calyx puberulus, raro glaber, 3-5-4 mm. 
longus; sepala ovato-lanceolata vel lanceolata, subacuminata. 
Corolla alba, suaveolens; tubus circiter 15-16 cm. longus, magis 
minusve puberulus, intus laxe pilosus; lobi rotundato-ovati, 
breviter apiculati, 2-2-5 mm. longi. Antherae 1-5 mm. longae. 
Stigma breve, obtusum. Bacca oblongo-ellipsoidea, 2-3-2-5 cm. 
longa, medio 1-5-1-7 cm. diametro, demum atro-purpurea vel 
exsiccando nigra, edulis. Semina ambitu late elliptica, circiter 
1-4 cm. longa, 1-1 em. lata, albida. A. venenata, Vatke ex 
Schweinf. in Engl. Jahrb. xvii. Beibl. 41, 46 (footnote), Holmes 
in Pharm. Journ. ser. 3, xxiv, 42; Stapf in Dyer, Fl. Trop. Afr. 
iv. 94 (partly). A. Schimperi, Schweinf. in Boll. Soc. Afr. 
Italia x. (1891) xi-xii, 13 (the Taita plant); and in Engl. 
Jahrb. xvii. l.c., Pax in Engl. Pfl. Ost. Afr. B. 519 (the Taita 
plant). A. abyssinica, K. Schum. in Engl. Pfl. Ost. Afr. A. 48 
aF 
AST Arrica. Kenya Colony; Taita, Ndara Mountains, 
1270-1525 m., Hildebrandt; Holmwood. Nairobi, common on the 
edges and in open places of the forests, 1780 m., C. F. Elliott 266; 
Battiscombe ; Kikuyu, Whyte. Usambara; Kwa Mshuza, 1300 m., 
Holst 8968. ‘‘ Deutsch-Ostafrika ” without precise locality, 
Busse, 382. — 
A. venenata, G. Don, appears to be confined to South Africa 
and Matabeleland (by streams near Buluwayo, Rand, 572; Hope 
ountain, south of Buluwayo, Baines). 
According to Mr. Battiscombe the fruits are edible having a 
sweet taste and are readily devoured by birds. Mr. Battiscombe 
writes :—‘‘ Both Acocanthera Schimperi and A. longiflora are 
used indiscriminately for arrow-poison, but that derived from 
A. Schimperi is said to be more potent than A. longiflora; to 
the Wakamba natives who are chief users of the arrow poison 
both trees are known under the name of ‘Ki bai’ but 
A. longiflorais qualified as ‘ Ki bai chi ao ’ e.g. black Ki bai. Just 
now the former is in full blossom and is a beautiful sight.” 
I have adopted Endlicher’s spelling of Acocanthera in the 
place of the absurd and barbarous form Acokanthera, found in 
G. Don’s Generum Systema, which is evidently due to a printer’s 
error. Don himself gives the derivation of the name as “ from 
axkwkyn, acoce (two ce’s) a mucrone...... ” Another printer’s 
error of a similar nature has crept into the same ae 
thera,” etc. and there are further misprints on the same page, a 
clear proof of bad reading. Pfeifer, Nomenclator Botanicus, has 
also Acocanthera 
1666. Brachystelma brevipedicellatum, Turrill [Asclepiadaceae- 
Ceropegieae]; affinis B. Arnotit, Baker, sed pedicellis brevioribus 
floribus majoribus corolla haud reflexa coronae lobis truncatis 
in parte superiore atropurpureis praecipue differt.. 
