62
ILLUSTRATIONS OF INDIAN BOTANY.
who happens to find an accidental variation of plant, not coming within the very narrow lines of demarkation which separate such genera, the right of at once constituting it a new genus. An Ivy by having a narrower more contracted calyx limb becomes Paratropia, or by having its styles cohering and somewhat elongated becomes a Gilibertia, or if its petals cohere and fall like a calyptra, no unusual case, especially when dried specimens are under examination, it passes into Sciodaphyllum. Can any thing be more trifling ? can the authors themselves of such generic characters look at them, when thus analysed and contrasted, without smiling?
Panax though very nearly allied, yet having an ovary with only 2 or 3-cells has its limits dearly defined by a comparatively invariable and important point of structure, but Maralia does not seem distinct from it. Gastonia having twice as many stamens as petals is another good genus. The one-celled ovary of Anthrophyllum ( Blume) seems to point it out as another.
I urge with greater confidence the reduction of the above named genera as Wallich has already done so in his list, where we find no fewer than 42 species ranged under the genera Pa- nax and Hedera. Those species, I presume, having a 2 or 3-celled ovary being referred to the former, the remainder, those namely having the ovary 5 or more celled, to the latter. The calyx being a little more or a little less developed, the petals a little earlier or a little later decidu- ous, the styles a little longer or a little shorter or more or less cohering, merit not the dis- tinction of being taken into account as generic characters : and to multiply genera on such characters is, in my estimation, alike unphilosophical and unworthy of science.
The genus Miquelia Meisner, our Araliacea ? Kleinii, if really a member of this order will probably prove a good genus, but in the mean time, while the male plant only is known, it seems to associate too well with Wallich's Phytocrene to be admitted as distinct- Both are di- oicous, in both the perianth is 4 partite and valvular in aestivation, the stamens in both are 4, the filaments united at the base with a sterile pistillum (on this point our character in the Pro- dromus is erroneous), in both the anthers are fixed by their back near the base, dehiscing lon- gitudinally, in both the stem is twining, the leaves alternate, the flowers congregated on the apex of the peduncles, forming a capitulum in the one, and a simple umbel in the other, the base of the pedicels in both are furnished with a cup-shaped 4-cleft hairy bractea or squama, and lastly, the structure of the stem is the same in both ; the very curious woody plates so ad- mirably represented by Mr. Griffith in Wallich's Ph. gigantea being equally found here.
The differences then between the two being, so Tar as our present information goes, only specific not generic, 1 propose referring our Araliacea? Kleinii to Wallich's genus Phytocrene & position which the identity of the lignious structure even more than the similarity of the floral one tends to confirm ; that structure being as yet unknown beyond this genus. The order to which the genus belongs, yet remains to be determined. The circumstance of our having referred one species and Meisner another to Araliaceae seems strongly in favour of this order finally proving its station, but until the female plant has been discovered this rests on mere conjecture.
EXPLANATION OF PLATE 118.
Paratropia venulosa. W. and A.
1 . Flowering branch, natural size.
2. A flower opened to show the relative position of the petals and stamens.
3. A flower, the petals separated showing the stamens, ovary and s-tigmas.
4. A flower, the petals separated at the base, but still hanging on the apex of the stamens.
5. An umbel of fruit.
6. A berry cut transversely, 5 -celled, with one seed io each.
Class 4th—Epicorollae Corisantherae. — Juss.
Torus between the tube of the calyx and the ovary, and usually forming at its extremity a small disk (epigynous) on the summit of the ovary, which finally coheres entirely, or rarely only at its upper margin, with the calyx tube. Petals usually united into a gamopetalous corolla, or rarely free, inserted on the outside of the disk. Stamens inserted on the corolla. Anther 0 distinct.
