SPICILEGIA FLOE.E SINENSIS. 105 



the quaternary or quinary flowers are constant characters, — and 

 recognising Ampelos (as I would call V. vinifera and its allies) as 

 a sub-groux3 of equal value with Kalocissiis, we can, I believe, effect 

 a very natural arrangement of the species. Lawson's revision of 

 VitacecB in the ' Flora of British India ' is very unsatisfactory. 



8. Crotalaria chinensis, Linn. In montibus Pak-wan, supra 

 Cantonem, d. 7 Aug. 1866, leg. Sampson. 



9. Medicago lapjiacea, Lam. Mr. W. G. Smith has suggested 

 (* Journ. Bot.,' xiv., 53) that this plant may have been introduced 

 into Bedfordshire from China, "because one of the habitats of the 

 plant is the rice-fields of Hongkong." This is a very fanciful 

 hj'pothesis. The locality where Mr. Lamont found the plant is 

 in the immediate neighbourhood of the race-course, where horses 

 are exercised every day of the year, and where there are a grand 

 stand and stables. There can be no kind of doubt that it sprang 

 up casually from seeds conveyed in imported hay, and it has no 

 higher claim to be regarded as a native of Southern China than 

 Medicago lupulina, Trifolii spp. pll., Cice7' arietinum, Vicia hirsuta, 

 V. teti-asperma, or Lathyrus Aphaca, all of which I have myself 

 met with at various times in similar localities, but which are in no 

 way established as weeds. 



10. Indigofera trifoHata, Linn. Copiosam in coUibus herbosis 

 oirca Cantonem, Jul. 1867, detexit Sampson. 



11. Glycine (Leptocyamus) tomentosa, Benth. In agro Amoyensi, 

 Oct. 1857, ipse legi ; ad sinum Ta-lien-wan, ManchuriaB, Julio 

 1860, coll. R. Swmhoe. 



12. Glycine (Leptocyamus) tahacina, Benth. In graminosis 

 insulae Tai-tan, prope Amoy, m. Maio 1866, invenit Sampson. 

 Not hitherto recorded from China, though found in some of the 

 South Pacific isles. 



13. Ruhus triphyllus, Thbg. ,* ft. intermmtius , var. nov. Calycibus 

 ramulisque densissime rubiginoso-glanduloso-setosis,carpellis quam 

 in typo foi-tius reticulatis. Chinkiang, 1876. W. G. Stronach. 

 (Herb, propr. n. 19,999.) 



I have never yet seen Asiatic or Australian specimens of this 

 •widely- diffused species with the peculiar and dense glandular 

 setae of this plant, which are indeed more thickly set than in 

 E. phcenicolasim, Maxim. ! It offers a direct transition to that 

 species, to which, when I first saw the Chinkiang examples, I felt 

 sm*e, without comparison, they were referable. They agree better, 

 however, with li. triphylliis, in theh more lacmiate leaflets, with 

 veins impressed above, less attenuate sepals, and deep purple 

 petals. The termhial leaflet is wide and truncate at the base, as 

 in phcmicolasius, but this is also the case in North Chinese and 

 Manchurian specimens of triphylliis. They certainly lead to a 

 suspicion that, if R. strigcmifi, Mx., be indeed reducible to R. Idmus, 

 Linn., R. phcenicolasius may not unlikely prove to be an extreme 

 form of Pi. triphylliis. 



* On the n.-m«-nelatnre of this species, see Focke, ' Batographische Abhand 

 luiigpii,' in ' Abhandl. d. Naturwiss. Ver. zu Bremen,' -k Bd., 2 Heft, 170. 



P 



