470 MR. H. BOLUS'S CONTRIBUTIONS 
Decembri), copiose legi. In loco unico tantum vidi; nec in Herb. 
Gubern. C. B.S. exstat. Herb. meo No. 4848! 
This is a very distinet species, its nearest congener being C. 
aculeatum, Sw., from which it differs in its rootstock, leaves, lip, 
and the peculiar uniform deep chocolate-coloured flowers. 
Following Harvey and Bentham, I had removed C. aculeatum 
to Eulophia under the name of E. plicata (Linn. Soc. Journ. xix. 
p. 336). Bentham observes (Linn. Soc. Journ. xviii. p. 317), 
under the head of Cymbidium, * the two African specics referred to 
it by Lindley have been shown by Harvey to belong to Eulophia 
[this refers to C. tabulare, Sw., and C. aculeatum, Sw.]; but the 
C. Sandersonii, Harv., from South Africa, and an allied species 
from tropical Africa appear to be true Cymbidia.” The only re- 
ference of Harvey to this subject which I have been able to find, 
is in the * Genera of S.-African Plants,' ed. 2, p. 360, where he 
says :—‘‘ The Cape species referred to this genus by authors belong 
to Eulophia,” and then alludes to C. Sandersonii, of Natal, as à 
geuuine species. Harvey gives no reason for his statement, and 
it is inconsistent, as I shall presently show, with the generic limi- 
tations given in the same work, where Cymbidium is described as 
having a “labellum without spur," and Eulophia with *labellum 
spurred or saccate at base." In the ‘Genera Plantarum’ the 
same distinction is adopted, only that Eulophia is somewhat ex- 
tended to include a “ labellum .... in gibbum saccum vel calcar 
produetum." So far as regards other characters, habit, and 
general appearance, the two species C. aculeatum, Sw., and the 
present plant have at least as much claim to belong to Cymbidium 
as to Eulophia. They seem to form a connecting-link between 
the Cape species of both genera, although these genera have been 
placed by Bentham in distinct subtribes. On this point what 
Bentham says is instructive :—* The division of the Vandes into 
subtribes is ,. . . difficult, and as yet very vague in its results. 
Habit, and even geographical distribution, has often to be more 
relied upon than any absolute character”? (Linn. Soc. Journ. 
xviii. p. 316). 
But I have now, from many fresh specimens, satisfied myself 
that neither C. aculeatum, Sw., nor this plant has any approach 
to even a gibbosity at the base of the labellum. This is, in both, 
attached to the projecting mentum of the column, and the rouud- 
ing off of the almost right angle of attachment alone gives the 
slight appearance of gibbosity. When the labellum is detached, 
