Page:Illustrations of Indian Botany, Vol. 1.djvu/213

Rh

Of this genus we remark in the Prodromus page 86, "Dr. Wallich (list No. 3723) enumerates Ternstrœmia crenulata from the Madras herbarium, with which we are not acquainted."

In my herbarium there is a specimen so named, but with a doubt (?) to which the following memorandum is appended.

"Ternstrœmia? but albumen wanting—Berry dry, 2-celled, at length, 2 valved, loculicidal, ovules 2? in each cell, seeds, one ripe in each cell, pendulous, wingless, plane on the inner side convex on the outer, albumen wanting, cotyledons flat, foliaceous, radicle small, straight, superior, testa of the seeds thick like parchment."

Of this plant I have since got numerous specimens, but all in fruit, so that I am still unable to give any account of the flower, but further examination with better materials has enabled me to correct an error in the above memorandum. The plant is a considerable shrub, with the leaves nearly confined to the ends of the branches, below which, on the naked portion of the branch, the berries are congested. The leaves are oval, oblong, acute at both ends, terminating in a short abrupt acumen, quite entire, somewhat coriaceous and glabrous on both sides. The berries are sessile, globose, pointed with the short persistent base of the style, glabrous, 2-celled, with a solitary pendulous seed in each, as above described, but the seed in place of having a thick testa like parchment and being exalbuminous I now find has a very thin testa, closely adherent to a copious albumen, enclosing the embryo. Hence this plant only diifers from the genus, so far as we can learn from the fruit alone, in having a foliaceous not a terate curved embryo, a difference however, which may be found accompanied with other characters sufficient to remove it from the genus. A figure of this plant will- shortly be published in the Icones.

Of this genus I have one species from Ceylon, which appears new, so far at least as can be judged from the very imperfect characters by which the species generally of this genus are distinguished. One of its most prominent features is a thick coat of hair on the back of the petals.

C. lasiopetala R. W. arboreous, leaves from oval oblong obtuse, to obovate spathulate, tapering at the base, coriaceous, entire, glabrous; flowers axillary, solitary, pendulous; sepals 5, with 2 small bracteas at the base; petals 5, very hairy on the back: anthers sprinkled with erect (not retuse) hairs : fruit globose.

Ceylon.—Colonel Walker.

The station is not mentioned but judging from the appearance of the plant, I infer it is from the more elevated regions of the country.

I found at Courtallum what appears a new Peninsular species of this genus, but which may possibly be only a small leaved variety of G. obtusifolia, and saw another at Nuera Ella in Ceylon, of which I had previously received specimens from Colonel Walker. They may be thus distinguished

G. parvifolia R. W. leaves from elliptic to ovate lanceolate, bluntly acuminated, retuse or emarginate, acutely, crenately serrated, glabrous; flowers axillary solitary, sessile, confined to the ends of the branches.

Courtallum in dense jungles.

This is perhaps too nearly allied to G. obtusifolia, but differs in the leaves of this being lanceolate, tending to ovate; in that to obovate cuneate; in this the largest do not exceed an inch or inch and half in length, and about half the breadth, while in that, they are usually twice the size : in this they are acutely crenato-serrated, in that the serratures are scarcely perceptible, both are glabrous, and in both the flowers are subsessile.

G. Ceylanica R. W. arborious, glabrous, leaves sessile, quite entire, from elliptic to oblong lanceolate, acute, tipped with a small retuse glandular point: flowers axillary on short peduncles; petals 5, obcordate: stamens very numerous, anthers oblong; ovary hairy, stigmas 5, clavate.

This species is nearly allied in some points to obtusifolia, but I believe quite distinct.

I. Eurya Wightiana, male plant—natural size.—2. A female flower.—3. The same dissected, showing the ovary and styles with the corolla opened, to bring into view the union of the petals at the base.—4. A diagram, exhibiting the arrangement of the floral envelopes.—5. The ovary cut transversely, showing its 3 cells and inclosed ovules.—6. The same cut vertically, to show the superposed vertical ovules.—7. Full grown fruit— natural size.—8. The same magnified.—9. The same cut transversely.—10. A seed with its podosperm.—11. Male flower dissented.—12-13. Anthers, the latter showing an occasional union of filaments.—14. The calyx and rudimentary ovary, the petals and stamens removed to bring it into view—all more or less magnified with the exceptions specified.

1. Gordonia obtusa—natural size.—2. A dissected flower, the petals removed, and the stamens separated to show the superior ovary and their slight union at the base—only four sepals are seen in the figure while five is the regular number, but the fifth being extern! usually separates with the bractioles.—3. A diagram of the parts of the flower, but imperfect with reference to the external sepal and bractioles.—4. Stamens back and front view.—5. Ovary cut transversely.—6. The same cut vertically, showing the ascending ovules.—7. The upper half of the ovary with style and stigmas. — 8. A fruit cut transverse]y.—9. A seed — natural size.—10. The same magnified, showing the wing-like appendage.—11. The same cut tranversely — with the exceptions mentioned all more or less magnified.