MR. G. BEtfTIIAM ON OECHlDEiE. 285 



Areihusece. Anther operculate, over the rostellum. 

 Neottieae. Anther erect, behind the rostellum. 



*** Abnormal tribes. 

 CypripediecB. Anthers 2. 

 Apostasies. Anthers 2 or 3 ; ovary 3-eelled. 



The primary division, founded on the consistence of the pollen, 

 has not been replaced by any other equally good, although it is 

 by no means absolute. The waxy pollen-masses of some species 

 of Phaius and JBletia, for instance, appear to be tardily formed, 

 the granular mass of pollen sometimes filling the whole anther- 

 cells ; the powdery pollen of Eriochilus, Acianthus, and some 

 others is almost consolidated into waxy masses ; and the waxy 

 masses of Earina and others will at length resolve themselves 

 into powdery granules ; but these exceptions are very rare, and 

 almost isolated among the immense number of genera where the 

 distinction is constant. 



The distinctions, however, founded upon the so-called caudicles 

 and gland can scarcely be maintained, independently of the con- 

 fusion occasioned by the term caudicle having been applied to 

 three very different parts of the pollinary system : — 1. The true 

 caudicle is the extension of the smaller end of a pollen-mass into 

 a tail-like point, corresponding to the caudicle of the pollen-mass 

 in Asclepiadeax It is specially exemplified in most Ophrydece, 

 and to a certain degree in a few other genera, such as Liparis, 

 JSria, CalantJie, &c. It is a part of the pollen-mass, though 

 often of a rather different consistence, and is included with it 

 in the as yet unopened anther-cell. 2. The so-called caudicle of 

 Epidendrum and its allies is, in like manner, included in the 

 anther-cell before it opens, but does not form part of each dis- 

 tinct pollen-mass. It is a variously shaped mass of loosely 

 connected pollen-grains, as variously attached to the two or four 

 pollen-masses of each cell, to which it forms a sort of appendage, 

 and might therefore, in technical descriptions, be distinguished 

 from the caudicle by the term appendicula, which seems more 

 appropriate than that of caudicula spuria given it by Blume. It 

 may be sometimes so much reduced as to make its presence or 

 absence very difficult to ascertain from dried specimens, and to 

 have caused several genera to be alternately placed in Malaxidece 

 and in Epidendrece ; in other, often closely allied, genera it may 



