A REVISION OF ACRIDOCARPUS 197 
and agrees ee Jussien’s phe 508 of the latter, as far as it goes; 
and since sieu ha n Thonning’s type, we are entitled to 
assume, in the absence - ‘ok any evidence to the contrary, that his 
identification of Vahl’s specimen ae? M. alternifolia was correct. 
now have, therefore, five points in favour of the identity of 
A. guineensis with A. corymbosus and siete its being conspecific 
‘ man j 
The only other species of deridocarpus which might, with some 
_ show of reason, be identified with 4. etter is A. longifolius 
Hook. fil. ; it differs, however, from both sieu’s and Thonning’s 
pir et a and it cannot, therefore, be conspecific with A. 
guineens 
A, tongifoin has perhaps been more misunderstood than any 
other species of Acridocarpus. Hooker fil. identified Fernando Po 
specimens of A. longifolius with A. guineensis Juss.; Oliver reduced 
it to A. Smeathmanni; and Engler has described different specimens 
of it, saa eth as @ new species, A. brevipetiolatus, and as a new. 
ee . Smeathmanni var. Dusenit 
Hooker does not state wherein “ A, guineensis’ , aifiars from 
A. long ie ; 4s comneria of his ae esi the two species 
yie daa. however, the following points of differen 
A, longifolius. ‘4, guineensis.”’ 
1. Perfectly glabrous. Uppermost parts puberulous. 
2. Branches slender. Branches stout. 
8. Leaves linear-oblong, mem- Leaves linear-lanceolate, cori- 
ee. ibs on the aceous, smooth on the upper 
upper surface. 
4, Racemes shutiina i Racemes lateral. 
(St. Thomas) ty Quint as, has the Seman parts just as be 
as in ‘* A. guineensis.” The fourth distinction is broken down by 
the fact that terminal and lateral racemes sometimes goo on ne 
Ww 
remaining diff 
different individuals of the same species, or even on the same 
individual, on parts of different ge. The Kew series of specimens 
of A. longifolius exhibits num tg ee intermediates in the shape, 
x 
represented by Mann’s Gaboon ive plant and Vogel’s No. 195 
m Fernando Po. The differences in the thickness of the 
branches are a slight. 
Let ow consider Oliver's reduction of 4A. longifolius to 
A, Siiaibwone. Ole a two varieties of 4. Smeathmanni, 
var. a with the wing of the fruit narrowed to the nut, and var. 8 
with the base of ‘the e wing half clasping the nut. Though Oliver 
does not state so, an examination of the specimens quoted a him 
