FLOEA or THE ISLAND OF HONGKONG. 393 
prove a good species, different from Q. purpure^m, Linn., so much the more as it seems that the latter has 
the inner hermaphrodite flowers in a greater number than our supposed variotv._Bj better essential cha- 
racters our plant .s distinguished from G. Udicum, Linn. (De Cand. Prodr. vol vi. p. 231. n. G2, and from 
^. i\ ^hacun^, Eaddi (De Cand. 1. c. n. 61, closely allied to it, and perhaps not different), both of which it 
at first sight resembles ; but they differ by their more slender habit and smaller size, by having narrower 
leaves attenuated m a longer petiole, whose enlarged base is somewhat auriculate-clasping, by smaller 
capitula and flowers, by the scales of the involucre being quite rufescent, and not tinged with green, and 
particularly by its very deciduous pappus, whose bristles are falling off always isolated, whilst in Q.-pwt. 
purmm, L., and our variety, the bristles of the pappus, also deciduous, are, when falling off, a^^va^ s united 
at their very base by a small ring, and thus cohering together. 
321. Gynura Pseudo-china, De Cand. Prodr. vol. vi. p. 299. n. 7.-Var. pubescens, Benth. + . 
Victona Peak and elsewhere (Hinds, Champion, teste cl. Benth.). 
322. Gynura auriculata, De Cand. Prodr. vol. vi. p. 300. n. 15. 
Var. a. gJahrata, De Cand. 1. c. Victoria Peak and elsewhere (Hance !). 
Var. ^. puberula, De Cand. 1. c. Victoria Peak and elsewhere (Hance !). 
Hooker and Amott's Gynura hulhosa, in Bot. Beech, p. 194, De Cand. Prodr. I.e. n. 19, and Gynum 
ovalis, De Cand. Prodr. 1. c. n. 11, both not seen by De CandoUe, seem to be the same species, for in all 
collections from Canton or Hongkong compared I have met with it, and never found any other. Philippi 
gathered it in China (no exact locality recorded), my friend Siemssen near Canton. It is easily distin- 
guished by its upper leaves clasping the stem at the base of their petiole by means of a large auricle, and 
by the pretty reticulation (caused by dark purple lines) mentioned by De Candolle, 1. c. Thcv arc not so 
conspicuous in the puberulous form as in the glabrous one, but when "the pubescence wears off," as well 
remarked by Hooker and Arnott, 1. c. "the under side is elegantly marked with numerous slender waved 
purpUsh veins." I have not seen an authentic specimen of this plant from Mauritius, and therefore T am 
unable to say whether it be different from the Chinese one. The short diagnosis of De Candolle, 1. c, 
however agrees well with it. 
323. Emilia sonchifolia , De Cand. Prodr. vol. vi. p. 302. n. 1. In waste places (Hinds, teste 
cl. Benth.; Hance! Seemann!). 
324. LiGtTLARiA K^mpferi, Sicb. et Zucc. Fior. Japon. vol. i. p. 77. t. 35.— Wlprs. Rep. Bot. 
Syst. vol. ii. p. 651. n. l.—Senecio K<smpferi, De Cand. Prodr. vol. vi. p. 363. n. 118, et vol. vii. 
p. 301. Hongkong (Hance!). 
Zuccarini's instructive figure agrees well with our beautiful plant, in no respect different from the 
Japanese one. 
325. Senecio campylodes, De Cand. Prodr. vol. vi. p. 370. n. 160.— >S. Chinensis, De Cand. 
vol. vi. p. 363. n. 116?? — S. Ilindsii, Benth. in Lond. Joui-ri. Bot. vol. i. p. 488, et in Hook. Journ. 
Bot. and Kew Misc. vol. iv. p. 235. n. 29 ! — "Wlprs. Rep. Bot. Syst. vol. ii. p. 654. n. 10. Ra\-ine9 
of Victoria Peak (Hinds, Champion, teste cl. Benth.; Hance I). 
326. Senecio Stauiitonii, De Cand. Prodr. vol. vi. p. 370. n. 1 17. Hongkong (Champion, teste 
cl. Seem.; Hance !). 
3 D 
