14
THE JOURNAL OF INDIAN BOTANY.
ad apicem sensim angustiora. Pedunculi 15-24 cm., a glabri ; vaginae 5 cm., ore acutae. Oapitula 6 mm. lata, globosa, nigronivea. Recepfea- culum glabrum, Flores ferimeres. Flos : — petala externe visiblia, magnis nigris glandulis instrucfca. Flos S ; sepala ad spatham anfcice Assam connata. Petala gequalia. Anthrae albae. Plate 43.
Peninsular India; Pancbgani, N. Kanara. Horsley-Knnda In Cuddappa District.
Scapes and leaves rather flaccid. Leaves very short compared with the scapes, otherwise the plants with the habit in general of E . collinum Hook f. Heads rather small floral bracts black, but covered with white hairs. Receptacle glabrous. Flowers 3-merous, normal but anthers white.
Var. megalocephala, Oapitula valde majora, 8-12 mm. lata. ; pedunculi breviores, 8-12 cm. Plate 44.
Central Provinces, Mandla District, Peninsular India, Mysore, Nilgiris 5.000 ft.
Peduncles much shorter and heads larger, possibly a distinct species. In habit like E. Geoffreyi, of which it might almost be thought to be a white anthered variety.
See App. I. E. Ritchieanum and E. melaleucum.
47. E. breviscapon Koern. ; F. B. I. vi 575, No. 16 ; Ruhl.
No. 7B. Stem short. Leaveswith broad sheathing base, then suddenly narrowed to a long linear blade, about 1/6 in. by 8-10 in. Scapes a little longer. Heads 1/3-1/2 in. Involucre black, or the outer bracts pale, reflexed. Plate 45. /
Peninsular India ; N. Kanara, in streams, etc. at 2,000 ft.
48. E. rivulare Dalziel ; F.BJ. vi 580, No. 30 (E. Dalzeli) ; Ruhl No. 138. Stem a few inches or 0, thickly covered with the very numerous leaves. Leaves linear, barely 1/12 in. wide not much enlarged at the base, 4 to 9 in. long. Scapes numerous, to 13 in. in length. Involucral bracts pale, at length slightly reflexed. One male petal longer than the others. Plate 46.
Peninsular India ; on the Ghats, at Castle Rock, etc. in streams.
A close connection of the last species of which it may be regarded as the more pronounced aquatic form.
Koerniche l.c. p. 605, rejected Dalziel's name rimdare because of a plant of W est Africa so named previously by Don. But Ruhland identifying Don’s plant with a still earlier E. lati folium Sm. restored Dalziel’s name.
49. E. miserum Koern. (Wall. Cat. 6070 in Herb. Calc.!); r.B.I. vi 575, No. 19 ; Ruhl. No. 72. Stem 1/6 in. thick, up to lj in. long branched at the base and so tufted, clothed below by the nume- rous filiform dead leaves. Leaves sheathing at the base, but soon
