BY J. H. MAIDEN AND E. BETCHE. 365 



radiating with shoots sometimes up to 18 inches in length. 

 Flowers considerably smaller; sepals smaller; of a different shape, 

 deltoid. Anthers apparently larger; filaments shorter. Hairs 

 on the fruit with a large gland on the apex. Hairs on the sepals 

 quite different, cellular; with whorled branches, sometimes like 

 the antlers of a horn of a stag. Pedicels longer and terete, not 

 flat and broader at the base as in ^4. Gibbonsii." 



As far as the habit goes there is no difference between this 

 form and the type. The type is variable in habit; we have seen 

 it quite prostrate, forming patches 4 to 5 feet in diameter, and in 

 other localities again almost erect. With regard to the details of 

 the flowers, Mr. Bauerlen's observations are quite correct in the 

 main. The chief and only essential difference is in the fruit and 

 the persistent calyx-lobes. The shape of the fruit is about the 

 same, but in the type the edges are densely ciliate with long 

 white simple hairs, and the calyx is at least half as long as the 

 fruit, and in the variety the hairs on the edge are shorter and 

 tipped with a gland, and the calyx is much shorter, the lobes 

 broader and the hairs are branched. 



These characters seem to be constant, but the general appear- 

 ance, habit, leaves and inflorescence of the specimens are so much 

 like the typical A. Gibbonsii, that we do not share Mr. Bauerlen's 

 view, who regards it as a new species. 



We have to thank Mr. R. T. Baker for the communication of 

 the specimens. 



COMPOSITE. 



Helichrysum diosmifolium Less. 



Wallangarra (J. L. Boorman; Nov. '04). 



A tall shrub attaining about 5 feet in height, with lanceolate 

 leaves, generally rather above | inch long, yellowish underneath 

 and slightly woolly, apiculate, the margins slightly recurved. 

 Involucral bracts pale yellowish. Flowers about 10 in the head. 



The common H. diosmifolium is a very variable plant, but the 

 forms pass so gradually into each other that we cannot divide 

 them into named varieties. The Wallangarra specimen described 

 above is the most broad-leaved form we have met with. Bentham 



