ON THE DUMETOKUM GROUP OF RU'BI IN BRITAIN. 171 



dentate-serrate, toothing more lobate than diversifoUus, and coarser 

 and much less regular than concinnus ; terminal petiolule equalling 

 one-third of its leaflet, bearing 7 or 8 falcate prickles ; stipules linear- 

 lanceolate, pilose, riovvering-shoot stout and rather flexuous as 

 compared with diversifoUus, armed with fairly numerous thick but 

 rather short and tolerably equal declining or subpatent prickles, more 

 distinct and fewer than in diversifoUus ; rachis clothed with rather 

 dense but not conspicuous pubescence, and very short setae and aciculi ; 

 but the intermediate-sized prickles of diversifoUus are notably absent ; 

 leaflets ternate or often quinate, in character as the stem ones, but 

 narrower in form and more decidedly sublobate or doubly dentate- 

 serrate towards their tips, especially the terminal leaflets, all hair-coated 

 and green but hardly felted beneath, thick, coriaceous, and fairly flat 

 above ; panicle more thyrsoidal and open than in diversifoUus, less leafy 

 to the top, and the axillary branches longer and more corymbose ; 

 sepals ovate-cuspidate or ovate-attenuate, dense-felted within, erect- 

 patent, or faintly clasping the mature fruit ; petals broad-obovate, 

 wrinkled, overlapping, rose-coloured ; anthers and filaments rich 

 yellow ; styles greenish ; fruit of 10-12 equally ripened drupels, smaller, 

 less csesian, and less unequally matured than diversifoUus. 



The rose-coloured petals, equal prickles, want of intermediate arma- 

 ture, the thicker leaflets, the more double serrature of the leaflets 

 of stem and panicle, separate this without much difficulty from diver- 

 sifoUus. From conchmus (to which it is much more nearly allied) the 

 more declining and less notably subpatent prickles of the rachis, occur- 

 ring nearly half as many again upon any given inch of it, the com- 

 parative sparseness of the leaf-felting, the stouter habit and armature, 

 and larger, coarser, general size of the individual bushes, the shorter petio- 

 lule of the terminal leaflet and the consequently more decided imbrica- 

 tion of the intermediate leaflets, the more straggly panicle, and generally 

 untidy aspect of the plant, the more deeply and unequally serrate toothing 

 of the leaflets, the deeper rose, not light pink, of the petals, sufficiently 

 distinguish it. 



This is tlie intermedlus of my notes in the ' Middlesex Flora,' (p. 100) 

 and I have now little doubt that Professor Babington's tuberculatus 

 includes much, perhaps all, of this variety of dumetorum. I fancy it 

 is better to restrict an already published name, than to invent a new one, 

 so I have abandoned my name '^ intermedius." I had merely adopted 



N 2 



