MB. G. BENTHAM ON OBCHTDE.E. 319 



the coriaceous, leas prominently veined leaves. I find, also, in 

 the only flower retaining its pollinarium which T could examine, 

 the pollen-masses sessile on the back of the rostellum, the gland 

 cr scale not being yet detached. The slender filiform stipites 

 represented in Parish's original drawing, rather too thickly 

 delineated in Fitch's plate (Trans. Linn. Soc. xxv. t. 29), appear 

 to me to be the viscum which in Vandea? so firmly attaches the 

 pollen-masses to the gland or stipes, and which is here more than 

 usually abundant, and readily stretches into elastic threads. 



Bromlieadia, Lindl., two Malayan species, and Polystachya, 

 Hook., about thirty species, chiefly African, but extending also 

 into the Indo-Malayan region, and sparingly represented in 

 tropical America, are both somewhat anomalous genera, with the 

 inflorescence terminating leafy stems, and often branched, but 

 with floral characters connecting them rather with Cymbidiea? 

 than with any other subtribe. 



Subtribe 3. CmTOPoniE.i;. — I have here collected twenty-one 

 genera, whose general character is to have the prominent mentum 

 of Maxillarieae with the foliage and habit of Cymbidiea?, thus form- 

 ing a connecting link between those two subtribes, but with limits 

 not always quite so definite as could be wished, for there are 

 here and there species offering exceptions to one or other of the 

 characters. The first three genera are terrestrial, without fleshy 

 pseudobulbs, although the base of the leafy stems often thickens 

 into a hard tuber. The others are epiphytical, and usually, if not 

 always, pseudobulbous. The leaves are generally plicate or with 

 prominent parallel ribs, as in the two preceding subtribes, and the 

 flowering-scapes are leafless in all except Govenia. The only ex- 

 ception to the mentum is in Aganisia, which, however, is too 

 closely allied to Zygopetalum to be removed from the subtribe. 

 The pollinarium has generally the stipes much more developed 

 than in Cymbidiea?. In geographical distribution, one small, rather 

 anomalous genus, PlocogJottis, is exclusively Malayan, and a mo- 

 notypic one, Pteroglossispis, is endemic in tropical Africa ; the 

 remainder are all American, though the typical one, Cyrtopodium, 

 extends also to tropical Africa and Asia. 



The Malayan genus Plocoglottis, about eight species, has been 

 well illustrated by Blume, but its systematic position may be as 

 yet doubtful. The anther is more distinctly two-celled than is 

 usual in Vandea?, and the two filiform slender stipites to the pol- 

 linarium are exceptional in the subtribe. Yet it has appeared to 



