ON A CHINESE SCEEWPINE, 69 



brunneo, Btigmatibus 1-3 in superioribus drupis eximie reniformibus 

 medio Bulcatis stylos crassos columnares corneos nitentes 1 J iineam 

 altos coronantibus introrsum vergentibus in mediis et inlerioribus 

 stylorum depressorum varie difformium tortorum ac spinosorum 

 apicalem faciei inferioriB superficiem drupae centrum spectatis occu- 

 pantibus et tunc ssepius minus exacte reniformibus v, etiam linearibus 

 cum stylis omnino persistentibus, semine trigono-conico apice compla- 

 nato basi coma fibrosa instructo lateri trophospermio magno fibrose 

 leviter adhserente testa albida raphidibus feta. (Exsicc. n. 18394.) 



Setting aside the phalangial species, and Gaudichaud's obscure 

 Souleyetia, Mr. Kurz classes the simple- druped uniovulate Pandani 

 in divisions founded on the structure of the stigmas, and on the free 

 or connate filaments. The male inflorescence of the Chinese plant 

 remains to be discovered ; but a consideration of its fruit apparently 

 leads to the conclusion that the sii-'matic structure is not so invariable 

 or so trustworthy as could be wished, and suggests a doubc as to the 

 stability of those characters on which Mr. Kurz relies. In his section 

 Acrostigma, the stigmas are described as erect, simple, spinose, and 

 turned outwards; in Ryckia as turned inwards, seceding, usually 

 forked, erect, or depressed ; and in Microstigma as sessile, lunate, re- 

 niform, hippocrepiform or two-lobed. Mr. Kurz originally limited 

 Acrostigma to those species with persistent stigmas, but in his last re- 

 vision he has somewhat modified its circumscription, admitting into 

 it his P. helicopus (originally placed in Pychia), which he describes 

 with "stigmata secedentia," explaning this to mean "imabasi fra- 

 gilia," as contrasted with ** stigmata persistentia, i.e.^ nonnisi cum 

 pericarpio separanda," The styles of the Chinese plant are persistent, 

 in the usual acceptation of the term ; that is, they remain firmly 

 attached to the di^upe at its full maturity ; and, though they may 

 be detached by force, leaving a scar, so can portions of the epicarp 

 itself. All authors concur in ascribing to PancUyms sessile stigmas ; 

 but in the present plant the styles of the upper drupes of the syn- 

 carpium are as well developed as in Sparganium ;* and in the lower 

 ones, though often much deformed and depressed, the papillose stig- 

 matic surface is always perfectly well defined and distinguished from 

 the horny style itself. Is not this also the case with many of the Indian 

 species? In the form of the stigmas the Hongkong Screwpine agrees 

 with Microstigma (they are quite as markedly reniform in the upper 

 drupes as in Kurz's figure of P. {Keura) utilis, Bory)f whilst in the 

 depressed, flattened, often irregularly branched spinose styles of the 

 middle and lower ones, — not unlike the chelce of some crustacean, — it 

 agrees better with Ryckia. No Indian species, I believe, except P. 

 (Acrostigma) caricosus, Rumph., has granular-topped drupes; but 

 but they are depicted by Mr. Kurz as distinctly and densely papillose ; 

 whereas in the Hongkong plant, the drupe is only finely granulated or 

 muriculate, as if sprinkled with sand, the asperities in nowise conceal- 



* J. G. Agardh suggests (Theor. syst. plant., 18) that PandanacecB perhaps 

 constitute the " formse primigenitge Spadiciflorarum." To me they appear to re- 

 present a lower type of development than Aracece, which view is clearly con- 

 sistent with Agardh's notion. 



t Journ. Bot. v., t, 64, f. 1. 



