di GENUS Oe FN A. 
By B. P. Paitties, MA. D.se., F-L:8. 
Tee National Herbarium has on many occasions received specimens of Ochna from the 
Forest Department for identification, but found that the naming of the specie: in herbaria 
was unsatisfactory. Forest officers have suggested that more species are found in the 
South African forests than those generally accepted as occurring within the Union, and 
as some of the species are of economic value it. was pointed out by the Chief Conservator 
of Forests that the genus should be critically examined. With the permission of the Chief 
o the Diviston of Botan y the writer undertook this work. Without the cordial co-operation 
f the Forest Department this would not have been possible, and my appreciation is here 
essed of the willing assistance received from Mr. J. J. Kotze, B.A., B.Sc., the acting 
Research Officer at the time. To the various forest officers who have supplied specimens 
and information the writer is also deeply indebted, especially to Mr. A. J. O'Connor, the 
District Forest Officer at Doornboom, northern 'transvaal, who first ee out the 
existence of two distinct species of Ochna which occur in the Woodbush Forest, and which 
had previously been regarded as either 9. arborea or O. atropurpurea, var. natalitia. My 
thanks are also due to the Director of the South African Museum, who allowed me facilities 
for working at the Cape Government Herbarium, and to Dr. 5. Schonland, Mrs. F. Bolus, 
B.A., Mrs. R. Potts, and;Dr. P. J. van der Bijl for lending me the material in their 
respective herbaria, and to the Director of Kew for comparing some specimens with the Kew 
material. 
At first sight the division of the genus into species appeared an easy matter, but as more 
material was examined the difficulties of limiting the species increased. The writer has 
critically examined nearly 200 herbarium sheets, but failed to find any constant characters, 
either floral or vegetative, which could be used as diagnostic characters for a key to the 
species, and has had to rely on the general “ facies ` of the specimens in most cases and has 
found it difficult to express the differences in words. That the specimens have been grouped 
correctly may be assumed, as handling a large number of specimens makes one familiar 
with their “look,” but whether a particular specimen can be readily placed in its correct 
species with the aid of the key given hete is more doubtful. However, this difficulty will only 
apply to a few species related to O. atropurpurea ; the others are fairly distinct. 
In the “ Flora Capensis ” (Vol. 1, 448}, three species of Ochna are described, but my 
investigations show that there are nine distinct species in South Africa with the possibility 
of a tenth. This latter is represented by two sheets illustrating a plant with a prostrate 
habit, but as Harvey, in the ` Flora C apensis index @. arborea, mentions that `` shoots 
from the roots occasionally produce flowers,” it is just possible that the specimens may 
represent such shoots, but no information on this point is available. In all the herbarium 
specimens examined only O. pulchra was correctly distinguished, while the limits of 
O. aiborea and O. atropurpurea were confused. 0. pulchra i is a typical Transvaal species, 
but also occurs in Rhodesia. and as far as our records go is absent from the Cape Province 
and Natal. It is readily distinguished from all the other South African species by the large 
racemose inflorescence. Other distinguishing characters are the entire leaves (or they 
usually appear entire to the naked eye) and the long stigmas, which are one-third to a quarter 
the length of the style. The other species which has entire (or apparently so) leaves is 
0. arborea, but this does not occur in the Transvaal. It has been recorded from the George, 
