29G PLANTS OF MADEIRA. 



sparingly-branched root are eaten either boiled or raw. They are out- 

 wardly black, internally white and subfarinaceous, and being eagerly 

 sought after for food by the fishermen and Orchil-gatherers resorting 

 to the Great Dezerta, when prevented by bad weather from procuring 

 better pronsions from Madeira^ there is reason to apprehend that the 

 plant will soon become extirpated. Already it is rare ; and it was only 

 in one place that I succeeded in obtaining a distinct sight of it, grow- 

 ing out of fissures, or on ledges, far down the face of the perpendi- 

 cular cliff, 1200 or 1500 feet high, which forms the eastern sea-wall 

 of the Great Dezerta, 200 feet or more below the edge. It can only 

 be gathered by expert pragsmen, let down by ropes from the top of 

 the lofty cliff-barriers which gird the island. It appears to flower early 

 in the spring. In June the flowers were all over, and the seeds nearly 

 or quite ripe. The unboiled root tastes like the tuber of Biiniim de- 

 nudatum, DC. ; when boiled it appeared stringy and insipid, like a bad 

 parsnip. It is much more dry, hard, and fibrous, than a carrot. The 

 Portuguese however on the spot call it Rock Carrot, "Cenoula da 



Rocha." 



12. Chrysanthemum Jicjematomma, Lowe. — C.fruticosum glabrum parce 

 ramosum, foliis succulentis rigidiusculis crassiusculis bipinnatifidis, 

 pinnis inaequaliter inciso-dentatis basi utrinque 3-5-pectinato-aen- 

 tatis, floribus subsolitariis paucisve (2-3) terminalibus in corymbum 

 amplum digestis, anthodiis crassis carnosis, radio pallide roseo v. 

 carneo, disco atro-sanguineo mox conico-convexo, 

 Hab. In rupibus maritimis Insularum Desertarum illius praesertim 

 Australis Bugio dictse, 



A genuine ChrysantJtemnm of De CandoUe's sixth Group, Magarsa^ 

 notwithstanding the remarkably convex disc of the receptacle, which 

 becomes hollow and more conical as the seeds ripen. Nothing can 

 exceed the beauty of this fine plant as seen by Mr. Wollaston and 

 myself early in June, 1855, on its native rocks of the Bugio, with its 

 masses of large pink flowers, varying from full rose to the faintest 

 blush or almost pure white, visible a long way off on the high perpen- 

 dicular barren cliffs towards the summit of the island. Even the 

 white-flowered state is at once distinguishable from its nearest aUies, 

 C. pinnatijidumy Linn, fil., and C. dissectum, Lowe (Novit. p. 17 or 

 539), of Madeira, by the dark blood-coloured florets of the disc, with- 

 out recourse to the other characters of the leaves and habit. 



