Win. Griffin, Esq., of South Lambeth, by whom it was 
received from the Cape of Good Hope. Our drawing was 
made in his hot-house, in June last. 
It must be considered the type of the genus Disa, from 
which we believe it will be found necessary to separate 
D. comuta, and several other species with spiked flowers, 
on account of their simple clinandrium and horizontal 
anthers, in the nature of which they approach the curious 
form of Pterygodium. • 
If the columna and stigma of this genus be e xamined 
when the flower is expanded, it seems impossible that any 
communication can take place between the pollen and the 
stigma, on account of the fleshy dilatation of the clinan-. 
drium which is interposed. But if the flower is dissected 
in an early state, it will then be found that the two lobes 
of this dilated process are folded together, and inverted 
over the stigma in such a manner, by the contraction of' 
the columna, that the glands are applied immediately to 
the stigmatic surface. At this stage also of the flower, the 
grains of pollen, which ev,entually cohere in the form of 
an indefinite number of waxy, wedge-shaped bodies, are 
all separable by the aid of a little nitric acid, and appear 
to be in their greatest state of perfection. 
A bulbous-rooted orchideous plant, of extremely diffi- 
cult cultivation. It will probably succeed best in very 
fine sandy peat, never allowed to become saturated with 
moisture, and, during the period when the plant is at rest, 
kept quite dry. The stem is a foot high, clothed with 
spreading, lanceolate, acuminate leaves, sheathing at the 
base. The jiowers grow two or three together, and are 
very large and showy ; of the outer segments the two 
lower are oval-lanceolate, and bright crimson, the upper 
of a delicate pink colour, and furnished with a short 
conical spur behind. The two inner segments of the flower 
are very small, and placed at the base of the column ; 
while their place in the flower seems to be supplied by two 
petal-like appendages adhering to the column on each 
side. The Ai^d/t^in is smaU, and linear-lanceolate. 
J. L. 
