333 FLORA INDICA. [Berberidea. 



umbellate than in the following). The branches are not glaucous. We have seen 

 no J5tna specimens of this plant, which Moris describes as intermediate between S. 

 Cretica and vulgaris, and adds that it retained its characters of habit for five years 

 when grown side by side with B. vidyaris in the Turin Botanical Garden. Philippi 

 also, in his account of the vegetation of JCtna (Comp. Bot. Mag. i. 92), states that 

 it is the same as B. vulgaris. There are also in Herb. Hook, specimens of this 

 species amongst Bourgeau's South Spanish plants, labelled B. .Sinensis, R.S., with 

 the synonym of B. vulgaris, var. australis, Boiss., appended by M. Cosson. These 

 specimens have the leaves less serrated than in the Himalayan form, but they are 

 very variable in this respect. Koyle's description of B. Kunawarensis is erroneous 

 in describing the panicle as leafy and pedicels as 3-5-flowered. 



6. brachybotrgs. In this, which is hardly distinguishable in many cases from var. 

 7, JEtnensis, the old leaves are very coriaceous. The flowers are abundantly produced . 

 The fruit in Kashmir specimens is large or small, reddish-black or covered with blue 

 bloom, on stiff and horizontal or pendulous pedicels of variable length. The Sikkim 

 specimens are extremely coriaceous-leaved, and some of them, not being in fruit, ai'e 

 perhaps referable to B. mnbellata or ^ax floribunda of B. aristata, which has brown 

 and polished (not glaucous) brauchlets. 



€. Cretica. The European state of this plant, from which we cannot distinguish our 

 Indian ones, has been described by Boissier as a southern variety of B. vulgaris in 

 the body of his ' Voyage Botanique dans le Midi de I'Espagne,' but in the appendix 

 he suspends his opinion in deference to Grisebach, who (Fi. Rumel.) says that it is 

 perfectly distinct. It appears under favourable conditions to gi'ow into B. cratnegina, 

 which again is not to be distinguished from wzY.Jloribunda of B. aristata except by 

 the fruit; indeed, Griffith's Bhotan specimens of B. aristata accord in habit and 

 foliage entirely with Ancher-Eloy's B. Cretica (391) from Libanus and the Grecian 

 Archipelago, having very small, nearly entire, lanceolate leaves, but difl'er in fruit 

 and the long pendulous raceme. Our Kunawar specimens accord perfectly with the 

 plate in Sibthorp's 'Flora Grasca.' 



3. B. aristata (DC. Syst. Veg. ii. 8); foliis valde coriaoeis ple- 

 rumque persistentibus obovatis oblongis lanceolatisve venosis varie 

 grosse spinuloso-serratis integerrimisve aoutis obtusis aristatisve ses- 

 silibus V. in petiolum angustatis, fioribua racemosis subpanioulatis v. 

 suboyraosis pendulis snberectisve, baccis stylo brevi stigmateque parvo 

 terminatis. 



a. normalis ; foliis amplis obovatis oblongis ellipticisve acutis 

 aristatisve (1-3-pollicaribus) apicem versus hie illie spinuloso-dentatis 

 utrinque viridibus v. subtus glaucis, racemis compositis multifloris v. 

 subeorymbosis, iloribus magnis, pedioellis rubris glaueisve. — DC. Prod. 

 i. 106; Hook. Exot. Mar. t. 98; Royle, III. 64; Wall. Cat. 14.74 et 

 1475 ex parte ! B. tinotoria, LescJi. in Mem. Mus. vs.. 306 ; iright et 

 Am. Prod. i. 16 ; Deless. Ic. Sel. ii. t. 3 ; Wight, III. i. i(. 8 ; Vanlioutte, 

 Flore dfs Serres, vi. ^. 7B ; Lindley et Paxton, M. Garden, i, 18. /. 5 ; 

 Wall. Cat. 1476! B. Chitria, Ham. mss. ; Ker in Bot. Reg. <. 739; 

 Bon, Prodr. 304. B. angustifolia, Poxh. Hort. Beng. 87. 



^.floribimda ; foliis obovatis oblongis lanceolatisve integerrimis v. 

 varie spinuloso-serratis subtus glaucis v. concoloribus, floribus racemosis 

 uoibellatisve, pedunculis ssepius elongatis simplioibus, pedicellis brevi- 

 usoulis elongatisve interdum valde glaucis. — 15. floribunda et B. petio- 

 laris. Wall. mss. sub 1474, et Don, Syst. Card. i. 115 ; Lindley in Penny 

 Cyclop, iv. 361. B. aristata, Wall. Herb. 1474, ex parte! B. affitiis 

 et B. ceratophylla, Bon, Syst. Card. B. coriaria, Boyle, mss.; Lindley 



