266 THE ORCHID REVIEW. 
and produce the gland at their apex ; yet the author says “ neither caudicles 
nor canals having any apparent attachment to a rostellum.” The gland of 
the pollinia is invariably rostellar in its origin, which would be impossible 
according to the new hypothesis. The idea of the fertile anther cell 
bearing on its surface an infertile cell or staminode is also due to the 
confluence of two distinct organs. The author, in a footnote, remarks :— 
“‘T have examined many flowers and drawings of species of Ophrydez in the 
hope of finding a species in which the infertile cells or staminodes are 
attached on the zmmer side of the fertile cells. But the only case in which I 
have found such an appearance is in Mr. Bolus’s drawing of Pterygodium 
carnosum, Lindl. (Orchids of the Cape Peninsula, t. 12, fig. 5). If the 
structures there represented as ovoid rugulose bodies really are staminodes, 
the fact affords a strong confirmation of my theory.” But, unfortunately 
for the theory, those bodies are the two stigmas. The fact is that the 
flowers of this group have undergone such an extraordinary amount of 
distortion, owing to the suppression and division of the different organs and 
the union of unlike parts with each other, that it is not safe to trust to 
appearances, and, to the extent pointed out above, the characters of the tribe 
Ophrydez in this work must be modified. And thesame may be said of the 
remarks under Diplomeris at page 338. It is only in the most highly 
specialised members of the group that this division of the anther and 
rostellum are seen to such perfection, and, by working backwards to the less 
modified species, every step of the development may be traced. A somewhat 
similar division of the anther is seen in Mystacidium and Dendrophylax, 
but no new theory is required to explain it. 
: R. A. ROLFE. 
LISSOCHILUS ARENARIUS. 
Tuis widely diffused and pretty species of Lissochilus has at last appeared 
in cultivation. Flowers which were produced in the collection of M. le Duc 
de Massa, Chateau de Franconville, France, where L. giganteus is said to 
flower each year, have been sent to Kew, by M. J. Sallier, of Neuilly. The 
tubers were introduced a few years ago from the Congo, by M. Dybowski. 
The sepals are green, slightly suffused with purple-brown, and acuminate, 
while the broad petals and lip are mauve-purple, the latter being rather 
darker in front of the two-horned crest, and the saccate spur yellow inside, 
with a few dusky-brown dots. It is widely diffused in East and West 
Tropical Africa, extending also to Natal and the Comoro Islands. 
R. A. Ro.uFe. 
