Page:Illustrations of Indian Botany, Vol. 1.djvu/249

Rh A large tree with opposite elliptic, oblong, coriaceous leaves, acute at both ends. Peduncles congested, axillary, 1 -flowered.

1. G. axillaris, (Blume). Java on the heights of Salak, flowers October and May.

This genus seems to form the connecting link between Guttiferœ and Hypericineœ, having the quaternary flowers' of the one and the polysporous cells of the ovarium of the other. It associates better with Garcinieœ than Calophyllieœ, on account of the ovules being attached to the axis, not the base of the cells as in that sub-order.

The species of this genus seem to be involved in considerable obscurity, apparently owing to no one having had an opportunity of comparing specimens from different countries. The original M. ferrea, is from Ceylon, and is well described by Linnæus in his flora Zeylanica. He quotes Rheede Mai. 3 t. 53, as a synonym, and in that, it is my impression he is right, though the figure represents a specimen greatly exceeding, in the size of its leaves and flowers, those I have from Ceylon. This however, Choisv has separated from the Linnæan plant, and called it M. speciosa, a very appropriate name, which, until furnished with better materials I shall not attempt to disturb. He (Choisy) quotes as an authority for his M. ferrea, a figure of Rumphius 7 tab. 2. This figure I have not an opportunity of consulting as my copy is incomplete, in that part, but I have reason to believe the synonym erroneous, as I have specimens from Mergui, communicated by Mr. Griffith, which seem exactly to quadrate with his character, though quite distinct from the Ceylon plant. Roxburgh again describes under the same name a plant diifering from bath, and readily distinguished by having its flowers terminal, rarely axillary, " solitary or in pairs" a character which perfectly corresponds with specimens communicated to me by Dr. Wallich, under Roxburgh's name, M. ferrea. The M. ferrea again of our Prodromus appears different from all the others in the diminutive size of its leaves and flowers, and in wanting the thick coating of white bloom on the under surface of its leaves so conspicuous on the Ceylon plant. The M. ferrea, of Blume, I cannot with certainty refer to any of the above, unless perhaps to my Mergui plant, and that merely on account of the introduction of the words " pedunculis axillaribus" indicative of the presence of a conspicuous peduncle, which it has, while all the others have the flowers nearly sessile, or on very short peduncles. These various forms may be thus characterized and designated.

1.M. speciosa, (Choisy D. C. prod. I, page 562). Leaves very long, linear, lanceolate, acute : .flowers sub- sessile; petals roundish, regular: ripe fruit 4-seeded. Rheede, Hort. Mai. 3, t. 53, excluding all other synonyms.

Woods of Malabar.

2. M. ferrea, (Lin. sp. 734). Leaves lanceolate, acute at both ends, ending in a long tapering acumen, bright shining green above, beneath white, from a thickish coating of a pulverulent or scaly incrustation: flowers axillary, solitary, subsessile.

Ceylon, frequent.

A beautiful tree, with numerous large, yellow, fragrant flower*, decorating its slender pendulous ramuli : altogether, when in flower, one of the most ornamental trees I hare ever seen.

3. M. Roxburghii, (R. W.) Leaves lanceolate, shortly acuminate, acute, coriaceous, bright shining green above, glaucous beneath, flowers terminal, solitary, or paired (one from the axil of each terminal leaf,) short pedicelled, petals subunguieulate, obcordate, curled on the margin, " capsule about the size of a crab apple, nearly round with an acute point, 1 -celled, 1-4 seeded, the partition nearlv obliterated, 2-valved."

M. ferrea, Roxb. Fl. Ind. 2, page 605.

Bengal about Calcutta.

This is perhaps too closely allied to the former, the principal difference being in the flower bearing ramuli, of this having but one or at most two flowers confined to the apex, while in the other they occupy the axils of many pairs of leaves all along the branch. It has besides the appearance of being a much more rigid plant than the Ceylon one. The terminal solitary flowers mentioned by Roxburgh 30 years ago, is still the same in my specimens gathered last year, and probably from very different plants from those which he described, a circumstance affording considerable collateral support to the opinion that they are distinct species.

4. M. coromandelina, (R.W.) Leaves narrow lanceolate, ending in a long tapering blunt pointed acumen, bright shining green above, paler beneath, but scarcely glaucous : flowers axillary and terminal ; peduncles shorter than the petiols.

M. ferrea, W. and A. Prod.

Courtallum in woods.

A very handsome tree. The very diminutive, size of the leaves and flowers of this, as compared, with those of all the preceding, not less than the disappearance of the white glaucous crust so conspicuous on all the others, lead me to consider it distinct. The leaves which in these are from 5 to 7 inches long and 1¼ to l½ broad, do not in this exceed 2½ long by about ½ an inch in breadth, the flowers are small in proportion.

5. M. pedunculala, (R. W.) Leaves lanceolate, acute, somewhat abruptly acuminated, glaucous beneath : flowers axillary, large, frequently paired, peduncles about twice the length of the petiols.

M. ferrea, ? Choisy, D. C. Prod. Blume, Bijd.

Mergui. — Communicated by Wm. Griffith, Esq.

The long peduncled twin flowers of this species, readily distinguish it from all the preceding, and proclaim the tree a very beautiful one when in full flower.

The very peculiar venation of the leaves at once distinguishes this genus, almost without reference to the fructificatiou, but the discrimination of the species is by no means so easy, indeed the circumstance that marks them out as members of the same family, a strong family likeness, equally serves to render difficult their distinction from each other. They do however afford some useful sectional marks in the number of parts composing the floral envelopes. In all, these parts, whether we choose to call them calyx or corolla, are petaloid ; to avoid therefore the inconvenience that might arise from calling parts sepals, which if removed from the flower and laid side by side with others occupying the place of petals, could not by external characters, be distinguished, I shall group the whole under the intermediate term perianth, and subdivide the genus into sections, depending not on the number of sepals or petals, which in the estimation of some Botanists, seem, in this genus, to be convertible terms, but according to the number of leaves of the perianth, the first section, having 12, the