276 I'HJ^ NATUEALIST. 



iii. Beacteat^.. These two sections include only species not found in 

 Europe.^ iv. Cinnamomi^. The species in this section are almost all 

 strangers to this country, v. Pimpinellifoli^. " Stems with numerous 

 prickles or unarmed, peduncles without bracts, leaves oval or oblong, 

 sepals connivent, persistent, disk almost absent." If this section only 

 included the pimpinelUfolim it would be very good ; but is R. Sahinl, Woods, 

 in its right place here ? The grey-tomentose leaves (on both sides) and glan- 

 dulose beneath, and the pinnatif\d calyx-divisions, are sufficiently decisive 

 characters to remove this species far from pimjyinelUfolicB. The same may 

 be said of R. involuta, Smith, vi. Centifoli^. " Prickles unequal, mixed 

 with setse, leaves oblong or oval, rugose, disk fleshy, sepals pinnatifid." 

 In this section we find R. Gallica, K, and 2t. proviucialis, Ait., with R. 

 Damascena, Miller. This last species seems to me too widely different in 

 all its aspects, to be united in the same group with the two preceding, 

 vii. ViLLOSiE. " Stems upright, prickles straight, leaves oval or oblong 

 with divergent serrations, sepals connivent, persistent, disk fleshy." R. 

 spinuUfolia, Dematr., with leaves glandulose beneath and glabrous above is 

 not in its proper place here ; the same with R. alba, L., which with its 

 arcuate prickles, and leaves glabrous beneath, is nearer the Canines than 

 the Yilhsce. viii. Kubiginos^. " The numerous glands which cover the 

 inferior surface of the leaves, are, I think, sufficient to distinguish this 

 from all the other divisions." A very natural section, bearing upon it a 

 stamp by which it may easily be recognised at first sight. Lindley wrongly 

 places here R. cuspidata, Bieb., and R. pulverulenta, Bieb., and also makes 

 a blunder in admitting under the Rubiginos^, R. Montezuma, Humb. and 

 Bonp., which has leaves glabrous on both sides ! ix. Canin^e. " Prickles 

 equal, recurved, leaves oval without glands, serrations convergent, sepals 

 caducous, disk fleshy." This section presents some confusion in the 

 grouping of the species admitted by Lindley. Of the nine species com- 

 prised in this section, only two are European, R. canina, L., and R. nibri- 

 folia, Vill. But R. indica, L., R. microphylla, Roxb., and ii. chinensis, Jacq., 

 are surely strangely placed here. x. Systyl.e. " Styles united into a 

 column." A very natural section, xi. Banksian^ only includes species 

 foreign to Europe, and included in the section Chinenses of the Prodrome 

 of De Candolle. 



(To be continued J 



(6) As I am only considering European species, and principally those of France, 

 I pass over exotic species in silence. 



