362 



was seen by Bentham and is now in the National Herb, of 

 Victoria. This appears to me to be specifically distinct from 

 the eastern specimens and to belong to A. halocnemoides. It 

 was probably from this source that Bentham drew his state- 

 ment that all the flowers are bisexual. This is not true of 

 A. arbuscula, where, if 2 pistils are found in the triad, it is 

 quite an abnormal occurrence. The only specimens mentioned 

 by Bentham as having been collected by Brown are those 

 from Port Dalrymple, Tasmania, and the Tasmanian plant is 

 the one here described and figured as A . arbuscula. 



A. (?) pruinosum, Paulsen, I.e. 63, from Carnarvon, W. 

 Aust., is described without fruit. The spike has 8-17 articles, 

 and judging by the photograph, plate vi., fig. 3, it is ^4. 

 halocne moides. 



A. brachystachyum, Paulsen, I.e. 64, fig. 26; tab. vi., 

 fig. 4, is described as having 4-8 fertile articles, perianth 

 exserted, pericarp brown and hard; seedcoat not mentioned. 

 Also from Carnarvon, W. Aust., and possibly a poor specimen 

 of A . leiostachyum. 



A. bidens, Nees in PI. Preiss., i., 632 (1844-5). I have 

 not been able to come to any decision with regard to this 

 species. Nees described it from a specimen without fruits 

 collected on the banks of the Swan River. A cotype (Preiss, 

 No. 1261) lent me from the National Herb, of Victoria, is 

 figured on plate xxxiv. of this volume, in the hope that this 

 may be of some assistance to future investigators. It is in 

 such early flower that it is only possible to say that the flowers 

 are arranged in 3's and apparently all bisexual. I have also 

 seen another of the specimens quoted by Bentham below his 

 description of the plant as Salicornia bidens (Fl. Aust., v., 

 204) — "margin of salt lakes, north of Stirling Range, F. 

 Mueller, October, 1867." This specimen has rather stout 

 branches and no unusual lobing of the barren or fertile spikes. 

 It has one imperfect (broken) spike 40 mm. long, with 13 

 articles, and 2 or 3 shorter lateral spikes. All perianths and 

 fruit have fallen from the spike. There is nothing to prove 

 that it belongs to the same species as the Swan River specimen, 

 or that it is not a .4. halocnemoides. Bentham's description, 

 I.e., of some fruiting specimen which he believed to be S. 

 bidens, agrees very well with A. hedocneinoide* as regards 

 perianth and fruit. That species, it may be mentioned, 

 sometimes shows in its lower barren articles (but not in the 

 fertile ones) a close approximation to A. bidens. It seems 

 impossible to make any further progress until some botanist 

 re-discovers this acute-lobed plant on the banks of the Swan 

 River and then traces it to the fruiting stage. 



