Dr Graliam's Descri2)tion of New or Rare Plants. 381 



subulate, unequal, the outer in general nearly twice the length of the 

 involucre, and always more than twice as long as those in the centre, 

 spreading or divaricated, and after flowering erect, bending across 

 each other, collected as in the bud, and invested by the involucre, 

 which also becomes erect. Flowers handsome, many of them abor- 

 tive, (always so in the ray ?) Cali/x obsolete, segments minute points 

 on the outside of the filaments, and alternate with the petals. Co» 

 rolla lilac ; petals 5, nearly equal, spreading, obovato-subrotund, entire, 

 undulate, glabrous on their inner surface, paler on their outer, veins 

 obscure. Stamens 5 ; filaments erect, equal to the petals ; anthers bi- 

 locular, ovaL, incumbent, white, marked while in the bud by a purple 

 line along their edges, at which place they afterwards burst ; pollen 

 white. Germen inferior, cordato-kidney -shaped, flat, the commissure of 

 the seeds being in the shortest diameter, rugolose, pubescent, crowned 

 by a thin, colourless, spreading, entire, flat, membranous border above 

 the insertion of the petals, each lobe marked towards its inner edge by 

 a crescent-shaped rib, so that the two together inclose an ovate space, 

 extending from the laase to the apex of the fruit. Styles 2, diverging, 

 shorter than the filaments. Stigmata capitate. Fruit when ripe brown, 

 undulate, verrucose, seed considerably narrower than its covering. 



We received the seeds of this unusually beautiful umbelliferous plant from 

 Mr Fraser, colonial botanist, New South Wales, in November 1827, un- 

 der the generic name Brunonia. They were marked " native of the 

 Island of Baracha." 



The plants were raised in a cold frame, and have been in flower in the Bo- 

 tanic Garden during August. They appear to belong to the genus Tra~ 

 chymene of Rudge, but the root-leaves of the species figured by him are 

 more divided, and on much longer petioles than in any of our specimens ; 

 his plant is much smaller, the rays of the umbel much shorter, the pe- 

 tals of a very diflferent shape, and shorter than the stamens, and, above 

 all, the fruit is said to be subglobular, instead of flat, as with us. Yet 

 the habit is so much alike, that a fear of multiplying names, without a 

 certainty of a difference of species, had led me to adopt his specific name 

 with doubt, while I at the same time pointed out the above distinctions, 

 and thought it difficult to suppose that so very beautiful a flower should 

 have been so long overlooked, if it grows, as he states his to have done, 

 near Port Jackson. On showing the proof-sheet, however, to M. Al- 

 phonse De Candole, I was informed by him, that his father considers it 

 certainly distinct, and will call it T. ccerulea. Such authority confirmed 

 my own doubts, and I willingly adopted this designation. I have since 

 heard from Dr Hooker that ne too suspects Rudge's plani may be dis- 

 tinct from ours. 



Villarsia lacunosa ? 



Y.lacunosa; acaulis, foliis coriaceis, ovato-reniformibus, subpeltatis, cre- 

 natis, subtus concavis maculato-punctatis obsolete venosis, petiolis ra- 

 dicantibus floriferis, floribus fasciculatis, corollis lateribus glabris api- 

 cibus obtusis, crenulatis, calycibus acutis. 

 Villarsia lacunosa. Vent. Choix. p. 9. ? 



Villarsia aquatica, Gmel. Syst. Veg. 1. p. 44?.? — Rcenu ^ Seh. 4. p. 180.? 

 Menyanthes trachysperma, Mich. Fl. Bor. Amer. 1. p. 126. ? 

 Description — Without ttem. Leaves all radical, floating, ovate, deep- 

 ly cloven at the base, lobes little separated, crenated, subpeltate, upper 

 surface slightl v convex, bright green, veinless, lower slightly concave, paler, 

 obscurely marked with broad flattened veins, and many irregular red spots, 

 and innumerable points of the same colour ; petiole round, greatly elon- 

 gated (1-2 feet), about half an inch below the leaf bearing a fasciculus 

 of flowers, and a cluster of slender, rigid, conical tubers^ from which pro- 

 ceed other petioles, bearing flowers and roots in the same manner, and 

 these again others, in endless succession. These tubers frequently be- 

 come black, and decay, in which case another cluster is produced in con- 



