RUB I ACE ^. 283 



even form a false verticil at the axils of the leaves as in the Lahiece; ^ 

 2. in the proportions of the involucre the bracts of which are some- 

 times short, sometimes large foliaceous imbricate ; here green, there 

 coloured in various ways ; in some cases entire, in others much 

 divided and pinnatifid as, for example, in the herbaceous Uragogas of 

 tropical Africa, whose inflorescences are terminal and which have 

 been named Trichostachys ; ^ 3. in the dimensions of the calyx which 

 may be short, entire or dentate, or large, foliaceous, more or less 

 closely imbricate in prefloration ; 4. in their interpetiolar stipules 

 which are sometimes united, sometimes more or less free, entire or 

 ciliate, dentate, or sometimes large and coriaceous as in the African 

 species named Gamptopus,^ Other characters also very variable are 

 the form of the stigmatiferous divisions of the style which are more 

 or less wide, thick, united or independent, clothed with papillae, 

 proximate, exposed or reflexed ; and the form of the epigynous disk 

 which is simple or consisting of two lobes or of two glands superposed 

 to the ovarian cells, depressed 'or hemispherical, or conical and more 

 or less raised. Several species also have polygamous or dioecious 

 flowers. We shall also see that in this type the form of the putamens 

 and of the seeds may present very numerous variations as we pass 

 from one species to another. 



In the woody and generally larger Uragogas which have been more 

 especially ranged in the genus Psychotria, the inflorescences are also 

 sometimes in capitules of cymes and the involucres are formed of 

 imbricate bracts, ordinarily long and narrow, sometimes coloured. 

 They have been named Patahea^ If, on the other hand, while the 

 bracts remain large and here and there coloured, or become small 

 and green, the axes of the compound or mixed inflorescence are 

 more or less elongated, so that it becomes a simple or, more generally, 

 a ramified cluster of cymes, we have the true Psychotria of authors, 

 which may also present, in the perianth, in the disk, in the gynsecium, 

 all the variations observed among the Uragogas with sessile flowers. 

 Hence arise a multitude of sections, usually considered independent 

 genera, which are here passed in review : — 



J Sect. 2. AxUlares (B. H. Gen. ii. 128). 4 Aubl. Guian. i. 110, t. 43.— J. iJ/ew. Mus. 



2 Hook. f. Gen. ii. 128, n. 271. vi. 401 (part).— Lamk. III. t. 65.— DC. Prodr. 



3 IIoQK. F. Bot. Mag. t. 5765. iv. 537.— Endl. Gen. n. 3142. 



