DlCTVOGENS.] 



ROXBURGHIACE^. 



219 



Order LXXII. ROXBURGHIACEiE.— Roxburghworts. 



Roxburghiaceae, Wall. Plant. As. Rar. 3. 50. (1832) ; Lindl. Nixns, 23. (1833) ; Endl. Gen. p. 1.57. ; 

 Meisncr, Gen. p. 402. ; Griffith m C'alc. Journ. Nat. Hist. p. 143. 



Diagnosis. — Dktyogens lo'ith bisexual flowers, solitary simple many-seeded carpels, with 

 long stallced anatropal seeds, and a basal placenta. 



Twining shi-ubs with tuberous I'oots % Leaves reticulated and coriaceous, with parallel 

 secondary veins connecting several primai'y ribs. Flowers large and showy, solitary, 

 foetid. Perianth of 4 large petaloid 

 divisions. Stamens 4, hypogynous ; 

 anthers adnate, opening inwards, 

 pomted, with connectives projecting 

 far beyond the cells, which separate 

 from the latter as far as theu* bases. 

 Ovary superior, 1 -celled, with 2 po- 

 lyspermous placentce arising from the 

 very base of the pericarp ; style none ; 

 stigma somewhat pencil-shaped ; 

 ovules 00, anatropal. Pericarp 1- 

 celled, 2-valved, with 2 clusters of 

 seeds at the base. Seeds attached to 

 long cords covered with loose hairs 

 just below the seeds ; embryo taper, 

 in the axis of fleshy albumen, with the 

 plumule lying within a slit. 



Fig. CL. 

 The affinity of this singular genus is not sufficiently marked to enable botanists to 

 refer it to any known Natural Order : by Endlicher it is placed at the end of Sarsaparillas, 

 and there can in fact be no doubt about its relation to the Parids, which that bo- 

 tanist mcludes in the Smilaceous Order. 1, however, formerly regarded it as more nearly 

 alhed to Arads than to anything else, and Mr. Griffith has so far agreed with that 

 opmion as to consider it certamly one of the class of wliich Arads are the type : in 

 which he has apparently been influenced by the discovery of a sht on one side of the 

 embryo. But this character has lost its value ever smce the discovery by Adr. de 

 Jussieu that a sUt embryo is found very generally in Endogens ; and a diclmous spadiceous 

 inflorescence is indispensable to Arads ; so that this view of the affuuty of Roxburghia 

 can hardly be mamtamed. It would rather appear to be the tj-pe of an Order lor 

 recruits to which we have stfll to look. In the meanwhile it may be looked upon as 

 a tendency towards Arals on the part of Dictyogens. Roxburghia is said to have stems 

 100 fathoms long. Mr. Griffith regards the pistil as consisting, beyond all doubt, ot one 

 carpel only, as " is indicated by the obliquity of the ovary." Its double natm-e, as above 



Fig. CL.- Roxburghia gloriosoides ; 1. the fruit ; 2. a seed ; 3. the same divided longitudinally. 



