184 



CHINA. 



[Rosacea. 



Candolle attributes only three nerves to each segment of the leaf. Roxburgh's B. scandens (Fl. Ind. 2. p. 

 326,) has glabrous leaves and orbicular petals, and is now called B. macrostachya by Dr. Wallich, in his 

 List of East Indian Plants, n. 5774. 



In addition to the above species of this Order, Mr. Millett and Mr. Vachell have detected Crotalaria 

 cahjcina, Schrank, C. variegata, Wall., a very fine Tephrosia near T. Heyneana, Wall., Uvaria crinita, 

 DC, Dicsrma elegans, De Cand., Rhynchosia virgata, Wall. Cat., Desmodium triquetrum, DC, Alysicarpus 

 monilifer, DC, Lablab vulgaris, Savi, Pachyrhizus angulatus, Rich., P. trilobus, DC, Dalbergia scandens, 

 Roxb., and D. lati folia, Roxb., Arachis hypogea, L., Poinciana pulcherrima, L., Cassia Thora, L., 

 C. bicapsularis, L., (according- to Wall. Cat.,) and a new species ? and C. Fistula. 



Ord. XXX. ROSACEZE. Jass. 



1. Kerria Japonica. De Cand. Linn. Soc. Trans, v. 12. p. 156. 



We regret to say that the flower is double, as in all the specimens we have yet seen. 



2. Spiraea lanceolata; foliis oblongo-lanceolatis glaberrimis subtus glaucis obtusiusciile 

 serratis, corymbis umbelliformibus paucifloris glabris ad apices ramulorum brevium lateralium, 

 pedicellis gracilibus, staminibus uniserialibus, disco 10-partito. — Poir. Encycl. Meth. v. 7. 

 p. 354. Cambess. in Ann. Sc. Nat. v. 1. t. 25. De Cand. Prodr. v. 2. p. 542. — S. Canton- 

 iensis. Lour. Cochin, v. I. p. 394. — S. corymbosa. Boxb.Fl. Ind. v. 2. p. 512? 



S. corymbosa of Roxburgh, we believe to be the cultivated state of this plant, more especially as he 

 says, in his Flora Indica, that it is a native of China; he adds, however, that it is also a native of the Moun- 

 tains north of India, from which he seems to confound it with S. callosa, Thunb. (the S. Bella of Sims, 

 Bot. Mag. t. 242G.) The figure among his drawings, n. 949, at the India House, is not like either, having 

 neither the lateral few-tlowered slender umbels of the one, nor the corymbose panicle of the other. Mr. 

 Lindley has accordingly, in Wallich's List of E. I. Plants, n. 701. p. 21 and 248, considered it as probably a 

 variety of S. chamcedrifolia. 



1. Rubus parvifolius ; caule prostrato tereti, ramis tomentosis, aculeis numerosis recurvis, 

 foliis pinnato-trifoliolatis foliolis subrotundis basi cuneatis inciso-serratis imparl ssepe inciso- 

 lobato supra glabris viridibus subtus tomentosis, stipulis subulatis, floribus paucis laxe race- 

 mosis terminabbus, laciniis calycinis ovato-lanceolatis. — Linn. Sp. PI. p. 707. Lour. Cochin, 

 v. 2. p. 398. De Cand. Prodr. v. 2. p. 563.— R. tripbyllus. Thunb. Fl. Jap. p. 215. 



It is singular that Seringe, in De Candolle's Prodromus, and Mr. G. Don, in Miller's Dictionary, both 

 refer to R. parvifolius of Tlnmberg, when no such plant is described by him. We have, however, brought 

 hither as a synonym R. triphyllus of that Author, on the authority of a specimen from Nagasaki in Japan, in 

 Mr. Arnott's Herbarium, from Dr. Fischer of St. Petersburg, and agreeing minutely with Thunberg's 

 description. Linnseus's plant was given him by Osbeck, and is most probably therefore from China ; but he 

 and succeeding Botanists have surely erroneously referred to Rumphius, Herb. Amb. v. 5. t. 47. f. 1. This 

 last species has ovato-lanceolate leaves, and seems to be what Chamisso and Schlechtendal have described as 

 R. Tagallus, (Linnsea, v. 2. p. 9): if, however, it has tomentose leaves, but Rumphius does not say so, 

 it may rather form a species with what Thunberg calls R. Menus, but which can scarcely be the same with 

 the European plant. 



2. Rubus reflexus ; ramis teretibus rufo-tomentosis, aculeis parvis sparsis foliisque oblongo- 

 cordatis 3-5-lobis infra dense tomentosis lobo terminali elongate venis reticulatis numero- 

 sissimis, stipulis bracteisque lanatis fimbriatis, racemis spiciformibus interruptis folio plus 



