75 
hyaline, 2-nerved; Jodicules minute, cuneate-oblong; filaments shorter than the elongate 
anthers; ovary obovate, glabrous; styles long, their bases remote. 
Western HIMALAYA: in the Tons Valley, Tehri-Garhwdl, at elevations between 
3,000 and 5,000 feet, also up to 7,000 feet on the open hillsides from Lambatách to 
Bámsu and Kulni, on the northern side of the Tons river. At the higher elevations 
this plant forms compact masses of turf over considerable areas, and the peculiar 
purple tinge of its foliage produces a marked feature from a long distance. In the 
sandy bed of the Tons river it has been found in some abundance, endeavouring to 
maintain a foothold, 
This species is closely allied to Т. abyssinicus Fl. Brit. Ind. (not of Nees), from 
which it differs by its more compact habit of growth, its purple tinge, and by its 
shorter leaves and spikes, The spikelets are more crowded and contain fewer florets, 
usually only two. When first detected in 1897, in the Tons Valley, it was distributed 
under the name of Lepturus Roxburghianus, and this determination was accepted at the 
Royal Herbarium at Kew. А more careful examination of the structure of the florets, 
necessitated for the preparation of above description and the accompanying plate, 
raised a doubt in my mind as to the correctness of this name, even generically. 
The frequent presence of 3 florets in some of the spikelets, coupled with a habit 
of inflorescence resembling that of a Tripogon, were indications of a closer affinity 
to that genus than to  Lepíurus, I should also mention that through the kindnes- 
of Major Prain I had an opportunity of examining Roxburgh’s unpublished drawing 
of Rottbællia biflora, which in the Fl. Brit. Ind. is included under Zepiurus Roxburgh- 
їапиз Steud., and this confirmed my opinion in regarding these two grasses as generi- 
cally distinct. Specimens of the Tons Valley grass were therefore again sent to 
Kew with a note expressing my doubt as to the correctness of the former name, 
I received in reply a very interesting communication from Dr. Otto Stapf, who 
after a careful re-examination of the material representing Lepturus Roxburghianus in 
the Royal Herbarium, has very obligingly favoured me with his views on the 
subject. In the first place he is of opinion that Lepturus Roxburghianus should be 
regarded as an Orcpetium, and secondly that this genus should be removed from the 
tribe of Hordee to that of Chioridew, and next to Microchloa (see Flora Capensis, 
vii. 742, where this latter opinion is recorded by Dr. Stapf). The affinity of 
Rottbellia biflora (Lepturus Roxburghianus), as well as of  Oropetium Thomeum, to 
Microchioa had already been indicated by Roxburgh, who, in his “Flora Indica,” 
included all three under the genus &Rottbællia. In Dr. Stapfs opinion Lepturus 
Roaburghianug is represented at Kew by specimens collected hy Law in Bombay, and 
by King on Mt. Abu. They agree exactly with Roxburgh’s figure, except that in the 
former the florets are minutely silky-pubescent, whereas Roxburgh represents the 
florets of his plant as glabrous, A specimen collected by Mr. J. Н. Drummond in 
the Hissar district, and quoted under Lepiurus liozburghianus in the Flor. Brit. Ind. vii. 
365, has also pubescent florets, and the leaves and spikes are longer than іп 
Roxburgh’s plant, theugh quite similar in other respects. Regarding the probable 
position of Tripogon purpurascens in the genus, l take the liberty of quoting the 
following extract from Dr, Stupf’s note :— 
“ Tripogon, as it stands at present (it has not been worked up monographically), is a somewhat 
heterogeneous assemblage. The species on which the genus was founded is T. bromoidee, and this 
type is repeated in 7. filiformis and witn süght modifications іп T. capillatus, Wight and trifidus, 
Any. Вот. Bor. Garp. Оліс., Vor, IX. 
