586 Prain — A Review of the Genera 
more or less conjointly, but as Athroandra , in its original form, is the 
older of the two it is desirable to take it into consideration first. When, in 
1862, Hooker proposed the section Athroandra (Journ. Linn. Soc. Bot., vi. 
20) he pointed out that C. Mannii , Hook, f., the species on which the 
section was based, is ‘ probably generically distinct from Claoxylon , from 
all the species of which the perulate buds abundantly distinguish it ’. Along 
with C. Mannii Hooker described another species, C . Barteri , Hook. f. 
( 1 . c., 21 ad calc .), which he referred to the same section. Hooker at the 
time was unaware, and indeed had no means of knowing, that specimens of 
the same species had, in i860, been described by Baillon (Adansonia, i. 68) 
as Trewia ? africana} The suggestion of Baillon as to the generic position 
of this plant lends support, which is stronger for being indirect, to Hooker’s 
belief that Athroandra is generically distinct from Claoxylon . As outlined 
by Hooker, the section Athroandra is based on species which agree with 
Mueller’s section Enclaoxylon in having the male flowers with only inter- 
staminal glands, but differ from Enclaoxylon in having perulate buds, and 
in having entire in place of plumosely laciniate stigmas. To the section 
Athroandra Mueller subsequently added five species which share the same 
characters:— C. Welwitschianum , Muell. arg. (Journ. Bot., ii. 333); C. 
columnar e , Muell. arg. (Flora, xlvii. 437) ; C. membranaceum , Muell. 
arg. (Flora, xlvii. 437); C. angolense, Muell. arg. (Journ. Bot., ii. 333); 
and C. rivulare , Muell. arg. (Flora, xlvii. 518). The justice of Hooker’s 
original view that Athroandra is probably generically distinct is further 
confirmed by the circumstance that Engler has based on specimens 
of C. rivulare his genus Chloropatane (Bot. Jahrb., xxvi. 383), and that 
Wright has referred specimens of C. Welwitschianum to the same genus 
(FIor. Trop. Afr., vi. 1, 169), under the name Chloropatane Batesii. 
The section A throandra as understood by Mueller is, however, somewhat 
wider in its limits than the Athroandra outlined by Hooker or the Chloro¬ 
patane described by Engler and Wright. This is due to the fact that of 
the two characters, the perulate buds and the entire stigmas, which dis¬ 
tinguish Hooker’s section, Mueller has relied only upon the first and has 
treated the second as negligible. This has led him to include in Athroandra 
1 The oldest specimens of Trewia ? africana were collected in Sierra Leone by Afzelins. These 
specimens bear a manuscript generic name, the existence of which perhaps indicates that Afzelius, too, 
had felt that this plant is not congeneric with the plant on which Claoxylon , A. Juss., was subsequently 
based: prior to 1824 the basis of Claoxylon was regarded as an Acalypha. The specimens on 
which Baillon based Trewia? africana were collected in 1859 H Perrottet in Senegambia, on the 
banks of the Casamance, long subsequent to the establishment of the genus Claoxylon by A. Jussieu. 
The specimens on which Hooker based Claoxylon Barteri were collected by Barter in Southern 
Nigeria—in Lagos Island, at Eppah and in the Yoruba forests. By a typographical error the name 
Yoruba appears in the Linnean Society’s Journal as ‘Gomba’. Mueller, when he detected the 
identity of Claoxylon Barteri and Trewia ? africana , left this typographical error uncorrected and 
inadvertently introduced another by transferring the provenance of Perrottet’s specimens of Trezvia ? 
africana from the banks of the Casamance to the neighbourhood of the Niger. 
