AND CLEISTOGAMY IN ORCHIDS. 393 
is self-fertilized, and it possesses no rostellum. On referring to 
Blume's * Orchid. Ind. Archip.' (p. 181, t. 45. fig. 1, t. 48 D), I 
find he describes and figures S. australis under the name of 
Gyrostachys pusilla, with a distinct bifid rostellum like that of 
S. autumnalis, Rich. To reconcile these two opposing statements 
I examined a number of plants all referred to S. australis in the 
British Museum, and found that both authors were right. 
Specimens from Mussoorie, India, collected by Thomson, and 
Java by Lobb, have rostella like those figured by Blume; while 
the Australian plants collected by Caley and Robert Brown have 
no visible rostellum. And, what is also interesting, the specimens 
collected by H. O. Forbes in Timor in company with Diuris and 
Caladenia, typically Australian orchids, resemble the Australian 
species in this respect. Specimens from the Friendly Islands also 
agree with the Australian form. But the Japanese and Chinese 
plants are again different from both. 
I have no doubt that the whole of these forms are self- 
fertilized. The pollinia in the erostellate forms are so closely 
approximated to the stigma that it is easy to conceive how, as 
Fitzgerald shows, they speedily reach the stigmatic surface. In 
many flowers, on carefully opening them, I find the lip full of the 
loose friable pollen which falls abundantly into the hollow of the 
labellum which wraps the short thick column. I may add that, 
on examining the flower-spikes that are in fruit, it is seen that all 
the flowers are fertilized, which is not commonly the case where 
insect-fertilization is required. 
There seem to be four common methods of self-fertilization 
among Orchids :— 
1. By breaking up of the pollen-mass and the falling of the 
dust either directly upon the stigma or into the lip, whence it 
Comes into contact with the stigma. This, of course, can only 
happen in the case of orchids with pulverulent pollen, viz. 
Ophrydee and Neottice. It occurs in some species of Thelymitra, 
_ viz. T. nuda, T. longifolia, T. pauciflora, and in at least some 
forms of Spiranthes australis, Lindl. 
2. By the falling of the pollen-masses as & whole from the 
elinandrium into the stigma. This is probably not rare, but I 
have met with records of but few examples. In a flower of 
haius maculatus, Blume, I recently received I found that by 
shaking, the pollinia had entirely slipped from the ge 
21 
