320 THE JOURNAL OF INDIAN BOTANY. 



Peninsular India ; on the Western Ghats, Salsette, Khandala, 

 Castle Rock, Tirthahalli ! 



Stems tufted. Leaves glabrous, lanceolate or oblanceolate acute 

 or mucronate. 2-5 in. by 1/3 in. at widest. Scapes twice as high or 

 less, thickened below the very globose or truncate heads, which are 

 white by the thickly puberous exposed back of the floral bracts. Re- 

 ceptacle tall hairy. Sepals in both sexes 2 only ; deeply boafc-shaped 

 and thickened along the keel. Female petals'linear with long basal 

 hairs. Anthers black. 



A vary distinct species, allied probably to E. sexangulare L. but with 

 the keel of the female sepals less developed. The slight but distinct thicken- 

 ing of the peduncle just below the head is very striking. (See notes on E. 

 bombayanum Ruhl. and E. neesianum Koern. in Appendix I.) 



Other species belonging to this section are : — 



E. auslrale R. Br. (in Herb. Calc. "China"!) closely allied to E. sex- 

 angulare L., being similar in the head and the female sepals. 



E. alatum (in Herb. Calc. Coll. Col. Pirie. Cochin China ;) with glistening 

 ovoid heads and female sepals as in E. cuspidatum Dalz. 



There are also in Herb, Calc. two other sheets one Coll, Loher. No. 1602 

 in Phillipines is very similar to the E. alatum (above) but the wings of the 

 female sepals are coarsely toothed. The other has a label "ex herb hort, 

 Kew " and a number 1168, but no other identification mark. Two of the 

 female sepals are crested, one not deeply boat-shaped and not crested. 



Ruhland in his clavis VI has a small group of three Indian, fi ve African 

 and ona Australian species with the female sepals " dorsocristatae-alatae." 

 The Indian consist of (I) E. heterolepis, Steud referred to above, (2) E. pseude- 

 quinqiiangulare Ruhl. which also I cannot find but which he says is very 

 closely allied to E. heterolepis, and (3) E. trilobum, hut the Calcutta sheets of 

 this last species have certainly no crests to the sepals. In his clavis V. 

 (flowers reduced) E. bombayanum Ruhl. apparently belongs to this series, but 

 I have not seen it. In his Clavis I, (dimerous flowers) he has in addition to 

 E. longifolium Nees, several American species with winged female sepals, 

 one E. guianense Koern, being figured (p. 37). 



{To be Continued.) 



I am indebted to Messrs Blatter and Hallberg for the photograph 

 repi'oduced in plates 12, 17 and 39. 



