286 
THE JOURNAL OF BOTANY 
which hitherto it appears to have been confused) in foliage ; hut is 
readily distinguished by the flower heads, whose shape places it in 
another section (Pteropogon) of the genus, a placing confirmed by 
the fewness of the flowers. It appears to be more widely distributed 
than II. Manglesii, more particularly in the drier regions. Its 
flowering season is a little later, though it overlaps ; and I have seen 
a few presumptive hybrids. The following description is for purposes 
of comparison :— 
H. Manglesii F. M., capitulis hemisphsericis, bracteis petaloideis 
patentibus unguibus distincte cuneatis quam lamellae multo breviori- 
bus, lamellis late ovatis apicibus obtusis paucidentatis, corollis cam- 
pan ulatis. 
Lysiosepalum F. M. Fragm. i. 142 correctio descriptionis origin¬ 
al is : sepala, petala, stamina sex. As there is an epicalyx of three 
almost equalty-spaced bracts and the gynsecium is trimerous, the 
whole floral structure is practically actinomorpliic. 
The hexamery of perigone and stamens seems to be unique in the 
tribe Lasiopetalece , if not in the order. Attention was called to this 
very important character, hitherto strangely overlooked, through my 
specimen No. 782 sent to the Department of Botany with other 
examples of drier-area plants. On learning that this appeared to be 
an undescribed Thomasia, I carefully searched the whole of my 
Lasiopetalous material and found other specimens which appeared to 
be conspecific with 782 and with Lysiosepalum Barryanum F. M.; 
specimens of L. rugosum Benth. from Yorkrakine, Burracoppin, and 
Wongan Hills also proved to be hexamerous. It has since been 
ascertained that all the Lysiosepalum material in the herbaria 
of the Department of Agriculture, W.A., Department of Botany, 
British Museum, and National Herb., Melbourne (type-specimens) is 
likewise hexamerous. Herb. Sargent 782 differs from presumably 
normal L. Barryanum F. Muell. in its smaller leaves and flowers, the 
colour (greenish, instead of rufous) of the bracteoles, and in the 
indumentum of the peduncles, consisting of almost sessile stellate hairs 
intermixed with longer purple glandular (?) simple ones ; the petals 
are sometimes fewer than six or wanting. All these differences seem 
to me attributable to climatic and edaphic influences ; and at the 
utmost indicate only specific separation in view of the characteristic 
Lysiosepalous epicalyx, floral hexamery; and scale-clothed ovary. 
My normal (?) specimens come from Gnowangerup, and their pedun¬ 
cular indumentum consists of long-stalked stellate with a few com¬ 
paratively short purple simple hairs. In the fresh state both this and 
782 have a peculiar somewhat aromatic odour, reminiscent of newly- 
cut cucumber. 
[Sargent 782 was described by me (Journ. Bot. lix. 245) as 
Thomasia hexandra. While agreeing with Mr. Sargent in most of 
what he says, I cannot admit the conspecificity of T. hexandra and 
Lysiosepalum Barryanum, for, besides the much smaller flowers of 
the former, its bracteoles are smaller and narrower, the sepals dif¬ 
ferently shaped, the petals, prominent in L. Barryanum , are absent 
at least from all hexandra flowers examinee!, while the much shorter 
