ILLUSTRATIONS OF INDIAN BOTANY-
113
these sections almost every species of this large genus (containing about 200 species) can, for the most part, be readily made out.
EXPLANATION OF PLATE 136.
Campanula fulgens. Var. Capitula, flowering plant, natural size.
1. Detached flower and bract.
2. The same, corolla removed to show the insertion of the stamens and dilated base of their filaments.
3. A flower cut vertically, showing the relative position of all its parts.
4. Stamens, back and front views.
5. Upper half of the style and stigma, the latter covered with its spreading, rigid, collecting hairs.
6. Ovary cut transversely.
7. A capsule approaching maturity, with withered corolla still attached.
8. Cut transversely.
9. One cell opened vertically, style and stigma left to show the change of the latter after flowering.
10. A mature capsule after dehiscence.
11. Capsule of C. Alphonsii, after dehiscence.
12. Dehiscing capsule of Wahlenbergia Indica. All more or less magnified.
XC— GOODINOVIEAE.
This family, according to De Candolle's Prodromus, includes about ISO species. Two or three only, so far as I am aware, have yet been found in India, and a few in the Moluccas, nearly all the rest being natives of New Holland and the Australian Islands. One is a native of the Cape, and a few of the Coast of South America. As an Indian family, it is one of very secondary note, but is interesting in connection with the geography of plants.
Character of the Order. Tube of the calyx more or less adherent to the ovaiy: limb 4-5-lobed, entire or obsolete, persistent. Corolla gamopetalous, more or less irregular, tube split above, rarely 5-partible; limb 5-parted, 2- or rarely 1 -lipped, the middle lobes lanceolate, flat, the lateral ones thinner and more corolline; aestivation induplicate, rarely obsolete. Stamina united with the corolla, not with the style, alternate with its lobes; filaments distinct; anthers united or oftener free, continuous with the filaments, 2-celled, bursting longitudinally. Pollen simple or compound. Ovary 1-2- or 4-celled; ovules few or numerous. Style simple or rarely double. Stigma fleshy, surrounded with a cup-shaped membranaceous indusium, entire or 2- lobed, ciliate or naked. Fruit various, capsular, many-seeded with the septum, when present, usually parallel with the valves, or drupaceous, or nucamentaceous with definite seed. Seed erect, albuminous with a thick testa. Embryo straight, foliaceous; radicle inferior. Shrubby or herbaceous, not lactescent plants, variously hairy with alternate, exstipulate, simple, entire, dentate, or somewhat incised leaves. Flowers distinct.
Affinities. This family is nearly related to Campanulacece and Loheliacece from which it principally differs in the curious cup-shaped indusium of the stigma, and the want of milky juice. Lindley formerly divided the order, as established by Brown, by the elevation of the sub-order Scasvoleae, to the rank of a distinct order, under the name of Sccevolacece, a course in which he was not followed by any other Botanist, and which, on reconsideration, he has relin- quished, as they are again united in his "Vegetable Kingdom." On the nature of the stigmatic indusium, which forms so important a feature in this family, considerable difference of opinion exists between Drs. Brown and Lindley. The former asks, "Is this remarkable covering of the stigma in these families merely a process of the apex of the style? or is it a part of distinct origin though intimately cohering with the pistillem? To the latter of these hypotheses he seems inclined to give the preference, viewing it, as I understand him, as composed of a series of modified stamina. Lindley, on the other hand, regards it "as nothing more than a remarkable prolongation of the rim which surrounds the stigmatic surface of Heathworts and of the plates which cover the style of Crane's bills (geraniums) and Balsams. It is in fact the upper free extremity of the carpillary leaves, distinct from that prolongation of the placenta, which is named the style and stigma." Between two such authorities, it would savour of presumption my attempting to decide, but for myself, I think the latter the more simple and probable of the two explanations of its origin. The aestivation in this family is also peculiar. The edges of
