Page:The Flora of British India Vol 4.djvu/9

This page has been proofread, but needs to be validated.

FLORA OF BRITISH INDIA.

Order XCV. ASCLEPIADEÆ. (By J. D. Hooker.)

Herbs or shrubs, usually twining. Leaves opposite or obsolete, very rarely alternate, quite entire, exstipulate. Inflorescence various, usually an axillary umbelliform cyme; flowers regular, hermaphrodite, 5-merous. Calyx inferior, lobes or segments imbricate. Corolla lobes or segments valvate or overlapping to the right, very rarely to the left; tube or throat often with a ring of hairs, scales, or processes (the outer or corolline corona). Stamens at the base of the corolla, filaments free in Periploceæ with or without interposed glands; in other tribes, connate into a generally very short fleshy column, which usually bears a simple or compound ring or series of scales or processes (inner or staminal corona) that are attached to the filaments or to the back of the anthers, or to both; anthers crowning the column, connate or free, adnate by the connective to the stigma, 2-celled; tip often produced into an indexed membrane; pollen forming one or two granular or waxy masses in each cell, the masses united in pairs or fours to a gland (corpuscle) which lies on the stigma. Ovary of' two distinct superior carpels, enclosed within the staminal column; styles 2, short, uniting' in the stigma, which is 5-angled short and included between the anthers, or is produced beyond them into a long or short simple or 2-fid column; ovules many, rarely few, 2-seriate in each carpel. Fruit of 2 follicles. Seeds compressed, usually flat ovoid winged and surmounted with a dense long brush of hairs (coma) (absent in Sarcolobus); albumen copious, dense; embryo large; cotyledons flat, radicle short, inferior. — Distrib. Species about 1,000, chiefly tropical.

The analysis of the plants of this order is most difficult, and in dried specimens never satisfactory, from the fleshiness and complexity of the coronal processes and anthers. I have spent many months over the Indian ones, and have kept pretty close to the generic limits adopted in the "Genera Plantarum." I have, however, been obliged to abandon .the tribe Stapelieæ, to suppress Vincetoxicum, and to propose several new genera.

Suborder I. Periploceæ. Filaments usually free; anthers acuminate or with a terminal appendage; pollen-masses granular, in pairs in each cell.

Tribe I. Periploceæ. Characters of the Suborder.

* Coronal scales or processes 0.

Anthers with bearded appendages
1. Pentanura.

** Coronal scales corolline, free, short, thick.

Corolla very small, rotate, lobes valvate
2. Hemidesmus.
vol. iv.
b