294 MR. G. BENTHAM ON" OltCHIDE^. 



of the pollen-masses. Whether this persistence is really constant 

 in all the species remains to be proved ; the flowers are generally 

 so minute that their accurate observation in dried specimens is 

 exceedingly difficult. Similar anthers have only been observed in 

 Sunipia, an East-Indian pseudobulbous epiphyte with the habit 

 nearly of some species of Bulbophyllum, but which, on account of 

 this very important peculiarity in the anther, one cannot help as- 

 sociating with Malaxis and Microstylis in an artificial but distinct 

 subtribe. For this subtribe I have taken the name of Micro - 

 stylece, in preference to that of Malatcece, to prevent all confusion 

 with Lindley's tribe Malaxideae. 



Subtribe 3. Liparide^;.— The chief character of this subtribe 

 rests in the terminal inflorescence without the terminal leaf of 

 Pleurothalleae, and in the more or less distinctly 2-seriate pollen- 

 masses (usually four), which in the normal genera are either quite 

 free or with their points slightly connected after dehiscence by a 

 very small quantity of viscuin. The species are either terrestrial 

 or more or less epiphytical, mostly natives of the temperate 

 regions of the northern hemisphere or of the tropical Indo- 

 Australian region, very few being found in tropical America. 

 Of the eight genera we would include in the subtribe, the two prin- 

 cipal ones (Oberonia and Liparis), which have each about 50 species. 

 Eeichenbach unites the former with Malaxis, from which it appears 

 to me to differ as much in the structure of the flower as in habit. 

 The operculate incumbent anther is quite that of Liparis, from 

 which the generic distinction consists chiefly in the shortness of 

 the column and in a peculiar distichous foliage with usually 

 minute flowers in an almost spike-like dense inflorescence. It is 

 also more tropical in its geographical distribution, and limited to 

 the Indo-Australian and South Pacific regions. Liparis itself is 

 somewhat variable in habit, often assuming that of the Asiatic 

 species of Microstylis, from which the most remarkable deviations 

 are Thouars's section Dislichis, with its elegantly distichous bracts 

 and flowers, and the Andine L. ramosa, Poepp., with decumbent, 

 more or less branched, leafy stems. The generic name Liparis 

 was altered by the elder Eeichenbach into Sturmia and by Hoff- 

 mansegg into Alysia as having been previously in use among 

 entomologists ; but that objection is now no longer held as 

 tenable, and Eichard's name is universally adopted. I should 

 include in the genus not only Empusa, Lindl., already reduced to 

 it by Eeichenbach, but also that author's genus Ephippianthus 



