TTMBELLIFF.EM. 155 



hollow stem and fleshy orbicular, cordate, reniform or fan- shaped 

 leaves, dentate or lobed and covered with hairs. 



Some Asiatic or American Aralias, formerly named Aureliana ^ and 

 often also referred to Panax, "^ as A, tri folia, Ginseng (fig. 185, 186), 

 qiiinqucfolia, differ from the preceding only in the number of their 

 ovarian cells and styles, which is two, as inA.papyrifera. The floral 

 pedicels are distinctly articulate, and the top of the ovary is convex. 



In 1790 LouEEiRo distinguished as a genus under the name of 

 Plectronia,^ Aralia trifoliata,'^ a Chinese species whose petals are 

 valvate and pedicels ordinarily, but not constantly,^ without articula- 

 tion. It is scarcely possible to separate it from A, japonica. The 

 leaves, it is true, are trifoliolate and the gynsecium dimerous, but A, 

 pentaphjlla^ of Japan, which may have four or five ovarian cells, and 

 A. ricinifolia, with simple and palmatifid leaves, have been considered 

 congeners. The latter has also been referred to Brassaiopsis,"^ which 

 comprises A. scandens, glomerulata, &c., because these have styles, 

 longer or shorter, united in the greater part of their length. The 

 pedicels, like those of A, japonica, are not articulate. This is the 

 only character which distinguishes them from A. disperma, calycidata, 

 &c., types of a genus Macropanax,^ which thereby approaches nearer 

 Aralia proper. The albumen may, indeed, be ruminated in Macro- 

 panax and Brassaiopsis, but the fact is not constant, which detracts 

 much from its value. The characters derived from the styles have 

 scarcely a more absolute signification.^ Nor can we consider other- 

 wise than a section of Aralia, Pentapanax,^^ in which the floral pedicels 

 are articulate or rarely not articulate, the petals often imbricate or 

 more rarely valvate, and the styles short or elongate, but united in a 



1 Lafit. Mem. ... Ginseng (1718) c. tab.; flowers certainly had distinct or obscure arti- 

 Catesb. Carol. App. (1743) 16, 1. 16. — Araliastrmn culations. 



Vaill. Serm. 43. — Aralice sect. Ginseng B. H. ^ Thunb. Fl. Jap. 128. — Panax spinosa L. f. 



loc. cit. S/ippl. 441. — Acanthojpanax spino&um Miq. Ann. 



2 Seem. Journ. Bot. vi. 52. — Eupanax Torr. Mus. Lugd.-Bat. i. 10.— Seem. Journ. £ot. v. 

 et Gr. Fl. N.-Amer. i. 647. 238.— Hook. Fl. Ltd. ii. 726. 



^ Fl. Cochinch. 162 (not L.). — Acanthopanax 'Done, et Pl. loc. cit. 106. — C. Koch, 



Seem. Journ. Bot. v. 238.— Mia. Ann. Mus. Vochenschr. (1859) 364.— B. H. Gen. 945, n. 32. 



Lugd.-Bat. i. 10. — Xalopanax Miq. loc. cit. 16. — Seem, Journ. Bot. ii. 290. 



— Panax subgen. Acanthopanax Dcne. et Pl. ^ Miq. Fl. Ind.-Bat. i. p. i. 763 ; Ann. Mas. 



loc. cit. 105. Lugd.-Bat. i. 13. — Seem. Journ. Bot. ii. 293. — 



^ Meyen. Reis. ii. 432. — Zanthoxylon trifolia- Hook. Fl. Lnd. ii. 738. 



turn L. Spec. 1445. — Panax aculeatum Ait. Hort. ^ See H. Bn. Adansonia, xii. 147. 



Keiv. iii. 448. — P. Loiireirianum DC. Prodr. iii. lo Seem. Journ. Bot. hi. 294.— B. H. Gen. 936, 



252. — Acanthopanax aculeatum Seem. n. 5. — Hook Fl. Lnd. ii, 723. 



* In certain specimens from Formosa, several 



