BY BARON VON MUELLER. 163 



is not dehiscent transversely, but rupturing irregularly at last, as 

 normal for Ptilotus. 



Incidentally new localities of some congeners may here be 

 noted : — 



P. spatulatus, northward to the Lachlan-River (F.v.M.). 



P. incanus, Finke- River (Rev. H. Kempe). 



P. Beckeri, Kangaroo-Island (Prof. Tate). 



P. helipteroides, Gascoyne-River (Hon. J. Forrest), Finke- 



River (Rev. H. Kempe). 

 P. Drmnmondi, Gascoyne-River (Hon. J. Forrest). 

 P. parvifolius, Lake Eyre (Prof. Tate), Grey's Ranges 



(Rev. W. Webster). 

 P. conicus, Croker's Island (Foelsche). 

 P. sjncatus, King's Sound (Froggatt). 

 P. Fraseri, var. Schwartzii, near the Macdonell's Ranges 



(Rev. Mr. Schwartz). 



Stems to about one and a half feet high ; calyx twice as long 

 as the bracteoles, towards the middle red ; stamens partially 

 sterile, the interjacent minute membranes acuminate ; ovulary 

 glabrous ; fruit almost equilateral. These characteristics seem 

 to be sufficient for raising this supposed variety to specific rank. 



This is also an apt opportunity for pointing out that Professor 

 H. Barllon recorded some time ago the anthers of Polycnemum 

 as really unilocular, though apparently two-celled ; in a litho- 

 graphed drawing of Hemichroa pentandra. printed with numerous 

 others many ))ears since for " The Plants of Victoria," but 

 hitherto unissued, the anthers were also delineated already as 

 unilocular ; as may likewise be seen in the xylographic illustra- 

 tion of /'o^yoiem^/m /;i?n<a?MZrM7?i, published in the " Key to the 

 System of Victorian Plants," II., figure 34, as copied from the 

 above-mentioned lithogram. 



