140. ERIOCAULACE.E. [1. Eriocaulon. 



Peduncles mostly showing a spiral twist, their sheaths -S-'G" also 

 often twisted, membranous above. Heads button-shaped to ovoid, 

 •1 — 18"" diam. pale. Invol. bracts longer than the floral, ovate- 

 lanceolate to narrowly lanceolate acute or obtuse. Floral bracts 

 cuneate obovate or oblanceolate, sometimes slightly hairy at back 

 towards the minutely cuspidate tip. Receptacle distinctly villous, 

 columnar. Male calyx with 2 distinct sepals, sepals linear tapering 

 to the base, tipped with few hairs, corolla stipitate very minute 

 imperfect with usually one distinct linear petal shorter than the 

 stamens, pedicelled. Fem. fl. pedicelled ; sepals very slender or 0. 

 Petals 3-0, linear acute with few erect hairs, ovary usually slenderly 

 stipitate. Seeds broadly ellipsoid, with minute transverse areoles. 



Singbhum ! FL, Fr. Nov.-Dec. 



This is not typical achiton, which has very small rounded heads -I-'IS" diam., 

 with very rounded silvery membranous involucral bracts not longer than the 

 head, inner dark grey, female sepals 2-3 oblanceolate hairy, petals 3 linear narrowly 

 linear hairy or of 3-0 hairs. 



11. E. Edwardii, Fyson (Journal of Indian Botany, vols, ii and iii). 



The following is Fyson's description (transcribed) : L. 2-5 cm. long, 

 4-6 mm. broad at the base, acute, glabrous, slender. Peduncle 6-20 cm. 

 slender. Heads 3-5 mm. obconic but finally globose, involucre bracts 

 not reflexed, 4 mm. long, scarious, glabrous, elliptic, acute, a little 

 longer than the floral bracts. These glabrous, obovate acute, grey 

 (pallido-nigrse). Receptacle tall villous. Fem. fl, Sep. 2, slender, 

 petals 2 linear or 0. Si. fl. sepals 2 ? connate into a spathe, petals 0, 

 anthers 6, black. Heads globose or nearly so, in bud overtopped by 

 the acute involucral bracts. 



Ranchi {Clarke's No. 34327) ! 



From an examination of the same material I find that the base of the leaves is 

 somewhat hairy, the floral bracts sometimes minutely hairy. ]\I. sepals sometimes 

 2 quite distinct and lanceolate or linear-oblanceolate. Fem. petals often fihform, 

 sometimes 2 capillary and a third linear, so that I conclude the Singbhum plant 

 referred by me to achiton is almost intermediate between that species and tliis. 

 The involucral bracts are too short and obtuse to be true Edwardii, which species 

 however I would personally consider merely a form of achiton. 



Var. Clarkei. 



To this species I refer provisionally Clarke's No. 24788 collected at 

 Giridih (Hazaribagh), which was left unnamed hj Hooker forthe F.B.I. 

 as being too young. It appears to be a variety with very acutely 

 cuspidate* bracts, involucre bracts erect -12" as long as whole head, 

 about 6, oblong, pale, finely acuminate. Fl. bracts • \" fuscous, 

 almost awned. Fem. sep. 2, narrowly linear. Petals but stipes 

 sometimes apparently villous above the sepals ! Male sepals spatha- 

 ceous, sometimes 3-toothed ( but Fyson described his type as spathaceous 

 with a " ?," and I think it is either a matter of age, or position in the 

 head, as I found obviously 2 free sepals in the type), staminal column 

 clavate with about 3-5-stamens and a black gland, petals or 1 

 minute linear. 



In the villous stipes it appears allied to the Brazilian E. griseum. 



1071 



