(3) LEO FARMAR. CONTRIBUTIONS TO AUSTRAUAN AMARANTACEiE, 1087 



T. Rœi, far reinoved in the a Flora Australiensis » froin this section 

 has stellate hairs, which were evidently overlooked and it is allied to 

 T. astrolasium. 



Trichinium axillare, F. v. M., described originally from poor rnate- 

 rial now proves frorn Dr. Clementi's spécimens to be a reinarkably free- 

 fiowering species. The stems attain over a mètre in length and are 

 clothed the whole length with heads, axillary on short peduncles below 

 and in large loose corymbs above. It should be noted that the ovary is 

 hairy, not glabrous as staded in the description. 



Although many members of thèse two gênera are exceedingly pretty 

 with their pinkish « ever-lasting » heads, often enveloped in white wool 

 they do not seem to hâve attracted the attention of horticnlturists ; 

 they are practically non-existent in European gardens. ïhe only ones 

 that appear to bave been grown at ail are Trichinium Stirlinqi, T. exal- 

 tatum and the beautiful T. Manglesii, which still üourishes in a few 

 botanic gardens. 



If we accept the revised définition of Ptilotus and Trichinium as 

 explained above, the following changes become necessary : 



Trichinium Garlsoni. Sp. Moore in Journ. Linn. Soc. (Botany) XXX, 

 220 (1898). 



= Ptilotus Carlsoni, F. v. M. in Vict. Nat., V, 74 (1888). 



Ptilotus depressus, W. Y. Fitz, in Journ. W. Austr. Nat. Hist. Soc, 

 I, 33 (1904). 



= Trichinium depressum, Farmar. (Nom. nov.). 



Ptilotus Polakii F. v. M. in Wing's S. Se. Kec. II, 274 (1882). 

 = Trichinium Polakii, Diels in Engl. Jahrb.. 191 (1905). 



Ptilotus Fraseri var. Schivartzii, F. v. M. in Proc. Linn. Soc, N. S. 

 Wales, p. 163, Ser. 2, III, (1888); Ptilotus Schivartzii, Täte in Trans et 

 Proc. Roy. Soc. Austr., XII, 82 (1889). 



= Trichinium Schwartzii, Farmar. (Comb. nov.). 



I hâve not, as has been done by others. attributed Trichinium Carl- 

 soni, Trichinium Polaldi, and Ptilotus Schwartzii to Mueller as he 

 himself has not actually formed those combinations. 



Other réductions necessary for reasons specified in each case are : 



Alternanthera decipiens, Benth. Flor. Austr., V, 251 (1870). 

 = Ptilotus Hoodii, F. v. M. Frag., VIII, 232 (1874). 



Bentham was in érror as to the anthers being one-celled an the leaves 

 opposite, For the same reasons in the Genera Plantarum of Bentham 

 and Hooker, Alternanthera polycephala \s reduced to Ptilotus macrotri- 

 chus (= P. villosiflorus). 



Ptilotus macrotrichus, F. v. M. Frag., IV, 90 (1863). 



= Ptilotus villosiflorus, F. v. M. Frag., III, 125 (1862). 



Bentham suggested that the latter was only a young starved and 

 small-fiowered State of the former. An abundance of material of the 



