Page:Illustrations of Indian Botany, Vol. 1.djvu/339

ILLUSTRATIONS OF INDIAN BOTANY.

15.9 pect both Rlume and Roxburgh bave described the flowers of their respective genera, Evodia and Fagara, erroneously in making them perfect and bisexual. Of the former I cannot be certain, but can have no hesitation with reference to the latter Botanisl having now before me male flowers of his F. Rhetsa and F. triphylla, the ovary of which, though apparently perfect, yet wants ovules and is sterile. Such is probably also the case with all Blume's species of Evodia, which must in that case, be brought here. This is the more likely to be the case, as all the other species of Evodia are from New Holland.

Z. triphyllum : (Juss.) Unarmed, leaves opposite, palmately trifoliolate, leaflets oval, oblong, acuminated, somewhat unequal sided at the base, glabrous: peduncles axillary, longer than ti e petiols, corymbs large, spreading; flowers numerous, minute: carpels 1-4, spreading, one seeded. Seed globose, glossy black.

Fagar triphylla, Roxb. Evodia triphylla, D'C.

Z. (Langsdorfia) ovalifolium, R. W. Unarmed, shrub- by, leaves alternate, palmately 3 foliolate, leaflets oval, shortly acuminated, somewhat cuniate at the base, entire, glabrous on both sides: panicles axillary, longer than the leaves, contacted, the short side branches ending in small capitate clusters of flowers. Fruit about the size of a pea, 1-seeded. Seed globose, cotyledons thick and fleshv, .dbuinen thin, membranous.

Sh^a'jerry kills in flower, and fruit in August and September.

Z. (Langsdoifia) sepearinm — Rigid, ramous, armed with numerous straight prickles: leaves palmately 3

foliolate, short petioled, leaflets oval, oblong, obtusely acuminated, crenulate on the margin, coriaceous, gla- brous : panicles shorter than the leaves, contracted, speciibrm, few flowered. Fruit about the size of a pea, 1-seeded. Pulicat hills near Madras.

This species is evidently very closely allied to the preceding, but is yet so different in every part of its habit that I cannot think of uniting them.

Toddalea aculeata and T. floribunda. In our Prodromus we remark that between the speci- mens of my catalogue No. 532, and Wallich's T. flori- bunila,v/e could point out no character to separate them, unless the tubercles on the stem. The plant here figured is identical with the specimen alluded to, and has, as I have since found, the tubercled stem represented in Wallich's plate ; hence is identical with it also. I therefore think (hat this may be received as a good species, and beg that the name'on the plate may be changed from aculeata to floribunda.

1. Toddalia floribunda, Wallich — natural size.

2. An expanded flower.

3. Stamens.

4-5. Sections of ovaries, showing them both sterile, and that this is actually a unisexual genus.

6. A female flower.

7. The ovary, the petals removed.

8. cut transversely, 5-celled.

9. cut vertically, ovules 2 in each cell,

superposed.

10. A young fruit.

11. The same cut transversely.

1 2. A seed — natural size.

13. magnified.

14. cut transversely, showing the cotyledons

enclosed in a thin albumen.

The figures representing the female flower were taken from specimens of T. aculeata, those of the other being all male plants.

This sub-order has been in a great measure treated of under the preceding, little therefore remains for this place. The genera referred to it were formerly placed by DeCandolle in his tribe Connaracece of the order Terebinthacece. Farther examination has led to that, at the time he wrote, ill understood order being greatly modified and to the removal of several of his tribes and genera to form new and distinct families. Those referred here were then transferred to Rutaceae, and more recently were by Dr, Arnott established as a 2d sub-order of that extensive order, an arrangement, partially adopted by Meissner who considers them a distinct tribe of Zanthoxylaceae. They principally differ from their associates in having solitary ovules and drupacious or samaroid (winged) fruit. Dr. Arnott gives the following character.

Sub-order 2.— Jlilantheae (Am.) Flowers unisexual, regular. Calyx 4-5, divided. Petals 4-5, alternate with the sepals : aestivation between valvular and twisted. Stamens as many as the petals and alternating, or twice as many, about the same length: filaments inserted round the base of the torus, not arising from scales : anthers 2-celled, bursting longitudinally. Torus discoid, or 4-lobed and cup-shaped. Ovaries several, distinct : ovules solitary in each cell, pendulous, styles very short, distinct, or slightly cohering at the origin of the stigmas : stigmas (distinct, filiform, recurved, papillose. Carpels 3-5, distinct, indehiscent, drupacious or sama-