WHE GENUS. -OlEN LA 
Py ON HOFMEYR and i. P. Puruiips MA Doe BLES. 
Division of Botany, Pretoria. 
THE genus Olima was founded by Thunberg in 1799 (Roem. Arck.,” II, i, 4, 1799), and 
belongs to the natural order Lythrarieae. The genus contains imporiant forest trees, 
and was first dealt with in this connection by Sim in his “ Forests and Forest Flora,” p. 227. 
Sim only recognized one species, viz. O. cymosa, but divided this into three varieties, as 
follows :—- 
Var. latifolia, mostly in the Western Province. 
Var. zntermedia, found in the Midland Conservancy and sparingly elsewhere. 
Var. acuminata, the common form in the Eastern and Transkeian Conservancies. 
The examination of all the material in the South African herbaria has led us to regard 
var. latifolia and var. intermedia as belonging to the species O. cymosa, and var. acuminata 
as a distinct species which Klotzsch first named O. acuminata. The leaf characters which 
Sim gives for the first two varieties are not constant in the areas in which they are stated 
to occur, and we have failed to find even constant varietal characters. The var. acwminat.: 
which has again been raised to specific rank can easily be distinguished from O. cymosa 
by the shape of the petals and, to a lesser degree, by the shape of the leaves. In the former 
the petals are “linear-spathulate” (Pl. IJ, figs. 2, 3), while in the latter (O. cymosa) the petals 
are `` obovate-spathulate ` (Pl. I, figs. 2, 8). These characters, in conjunction with its 
distinct distribution, all point to the tree being quite distinct from O. cymosa. It is the plant 
which Sim has figured on Plate LX XII in his work cited above. 
In the “ Index Kewensis ”’ two other specific names are mentioned, viz. O. ternata, Gilg 
and O. micrantha, Dere. As no other descriptions of these were availablein South Africa, 
the Director of Kew was asked for assistance. Jn a letter from Kew received in January, 
1921, it is stated that “ the name O. ternata, Gilg is apparently not vet published but Gilg 
determined a specimen himself collected by Mrs. H. Hutton at Howisons Poort near 
‘rahamstown as this species.” There can be no doubt that this is O. cymosa. From the 
description of O. ternata, Decne., kindly sent to us by the Director of Kew, we also have no 
hesitation in confirming this to be O. cymosa. Decaisne founded his description on Burchell 
3592, which was not represented in any of the South African herbaria. Another species 
from Zululand and East Pondoland was named by Dr. Stapf as O. radiata but the name has 
not been published. As this name is now known to foresters, we do not propose to 
change it. 
The Curator of the Botanical Department of the British Museum was good enough 
to furnish a scrap of Welwitsch 991 collected at Huilla and stated to be O. cymosa, but an 
examination of this makes it verv doubtful whether 0. cymosa does occur in tropical 
Africa. This has now been confirmed by Kew, as the Director writes : “* Welwitsch 991 
is not represented at Kew, but reference to the British Museum material leads us to agree 
that it is distinct from O. cymosa. It very closely resembles O. usambarensis (Holst 9115).” 
Through the courtesy of the Forest Department, we have been able to give some notes 
on two of the species, O. cymosa and O. radiata. 
