36 THE ORCHID REVIEW. [FEBRUARY, 1922- 
extent that the pollinia are lifted into a position vertically above the stigma 
(fig. 3), and the rotation continues until they are finally deposited on its 
frontal surface (fig. 4). Fig. 5 shows a case in which the movement has 
occurred in an oblique plane. 
In O. apifera the action of gravity is sufficient to explain the curving 
downwards of the caudicle till the pollinia are poised in front of the stigma, 
and their oscillation through wind is quite enough to bring them into: 
contact with it. Self-fertilisation may therefore have been due in the first 
instance, so to speak, to a happy accident. In P. gracile the transfer of the 
pollinia from anther to stigma is effected by such an ingenious piece of 
mechanism that the evidence of intentional adaptation is equally clear both 
for cross and self-pollination. As in O. apifera the latter is subsidiary to the 
former, and only takes place if necessary. 
Dr. Rogers states (in lit.) that “the stalk of the pollinium in all the 
Prasophyliums is a true caudicle and not a stipes asin the Vandez. It is the 
product of the pollen-masses and not of the rostellum. The pollen in this 
genus is sectile arid the various elastic strands from each pollinium unite into 
a single organ (caudicle) on the anterior face of the two masses near their 
common apex ; thence it passes on to its connection with the ‘ viscid disc’ of 
an emarginate rostellum. Both anther-case and pollinia are at first vertical.” 
There are several remarkable points in the functional activities of the 
mechanism for the pollination of P. gracile. The bending back of the anther 
so as not to hinder the rotation of the pollinia is worth notice, as well as the 
peculiar position of the anther at the back of the stigma, and not on the 
summit of the column as in European Orchids. The rotation of the pollinia 
through an angle of more than 180° after removal by an insect is striking, 
not only on account af the magnitude of the angle, but also because (judging 
by Dr. Rogers’ excellent figures) ita 
ppears to bedue toa curling movement 
of the caudicle itself, and not to th 
fl e hygrometric shrinkage of a part of the 
viscid disc. Further, self-pollination is brought about by the rotation of the 
pollinia iv situ, in the same manner as after removal by insects. No movement 
of depression occurs in the pollinia of any European Orchid until after the 
viscid discs have been withdrawn from their pouches, and then only through 
an angle ofgo®. It would be interesting to know whether similar movements 
of tha pollinia in situ are known in other genera of the Orchidacez. 
It is remarkable that a case of alternative self- 
parallel with that of Ophrys apifera (which belongs 
occur in such a very widely separated genus belongi 
pollination is some ways 
to the Basitonz) should 
ng to the Acrotone. 
HOLorurix JOHNSTONI.—An East 
near Zomba by Sir H. H. Johnston, 
larger flowers and a five-lobed lip. 
Tropical African species, collected 
It is allied to H. condensata, but has 
