364 



S. Australia. Without date or locality but probably near 

 Cooper Creek (Howitt Expedition, 1861-2, in National Herb, 

 of Victoria) ; Mt. Parry, between Lake Torrens and Leigh 

 Creek (R. Tate, Sept., 1883, in Tate Herb.); Marree 

 (J. M. B., October, 1917). 



N. Territory. Henbury Station, Finke River (G. F. Hill, 

 March, 1911, in National Herb, of Victoria as Salicornia 

 drier ea). 



Queensland. Georgina River (E. W. Bick, Sept., 1910 > 

 in Queensland Herb, as Tecticornia cinerea). 



W. Australia. Lake Cowcowing (Max Koch, No. 1147 r 

 Sept., 1904, in National Herb, of Victoria as Salicornia 

 DonaldsomJ. 



The Howitt specimen is one of those on which Bentham 

 founded his Salicornia tenuis, conceiving them to be the male 

 plant of a dioecious species. He says: — "The specimens are 

 very few and I do not feel certain that the male and the 

 fruiting ones are correctly matched." He was doubtless 

 misled by the fact that Howitt' s specimens are in early flower, 

 at which stage, owing to the proterandrous character of the 

 tribe, the stamens are much more conspicuous than the pistils. 

 In the type specimen placed at my disposal by Professor 

 Ewart I was able to find pistils with the characteristic 

 hardening of the young pericarp in some of the central flowers. 

 Usually it is possible to distinguish the species, even without 

 flowers, by the conspicuous scarious margins of the barren 

 articles. In the long, almost straight and cylindrical seed, 

 tapering at the base, it differs from any other species in the 

 tribe with which I am acquainted. During the hardening of 

 the pericarp and rhachis of the spike the seed appears to 

 revolve on its own axis to the extent of one-quarter of the 

 circumference of the cavity, so that, where there are two seeds 

 in the article, they lie side by side (as shown in pi. xxxvi., 

 fig. 11), instead of in the normal position of back to back. To. 

 ascertain whether this change is constant. would require the 

 examination of more material than was to hand. Sometimes, 

 through abortion, only one seed remains in the ripe article; 

 this is also true of P. robusta. 



J. McDouall Stuart's specimens from the "north-west 

 interior of South Australia," which Bentham accepted as the 

 female plant of S. tenuis, are in fruit, and certainly belong to 

 the species here described and figured as Arthrocnemum 

 halocnemoidcs , Nees. As the branches of P. tenuis are the 

 more slender of the two, it seems the proper one to bear 

 Bentham's specific name. Judging from the localities quoted 

 by Bentham, both species occur in the Darling district of 

 N.S. Wales. 



