the Brh'tJ}} Fuel, with particular Defcrlpi ions of each Species. 129 



frequently bifid, reprefenting in miniature the fruit-bearing termina- 

 tions of Fiictts vejicuhfus. 



The figure of Gmelin is fo accurate, though it reprefents a fingle 

 branch only, and not an entire plant, that it cannot be miftaken ; 

 and his defcription confirms our afTertion of the want of air blad- 

 ders, as he fays, the veficles confift of a congeries of tubercles ap- 

 pearing to be full of feeds. 



We have never obferved the terminating fubulate leaves to be 

 wanting in the perfedt plants, though in old ones they frequently 

 arc, being broken off by the force of the waves ; we have there- 

 fore adopted this as part of the fpecific charader, to diflinguifli this 

 fpecies from abrotanifollusy which always has the branches terminated 

 by a multipartite leaf with obtufe fegments. It is farther diftin- 

 guiflied from abrotanlfollus by the want of veficles, by being branched 

 diredtly from the root, and by the branches and fubdivifions growing 

 without order; in the latter alfo the veficles and frudlification are 

 obfcrvable on the upper branches only, but in this the tubercles 

 arc to be found on all the fmall ramifications. It is diftinguifhable 

 iidmjibiofusf by its fmaller fize, its terminating fru6lification and 

 want of veficles ; and irovn fceniculaceus and concatenatus of Linnseus, 

 by its want of veficles and its black colour when dried. 



The quotation of Gmelin's accurate figure and defcription of this 

 plant in Syjl. Nat. and the erroneous fynonyms given by Gmelin him- 

 felf, ai'e a fufficient apology for Mr. Hudfon's fuppofing this plant 

 thefcenlculaceus of Linnaeus ; notwithflanding the mention of veficles 

 in the fpecific charader, which are totally wanting in this fpecies, 

 might have led him to fufped fome error. 



We have notdefcribed the root; the fpecimens which have fallen 

 under our obfervation not having any : but it is reafonable to con- 

 clude, that both this fpecies and abrotanlfollus adhere to the rocks by 



Vol. III. S fome 



