The margins of the stipules are also occasionally united, so as to form a little horn-shaped tube. I have met with instances of this kind in the common white clover, Trifolium repens, where on each side of the base of the petiole the stipules had the form just indicated. That the bracts also may assume this condition, may be inferred from the peculiar horn-like structures of Marcgraavia, which appear to originate from the union of the margins of the reflected leaf.
Tubular petals occur normally in some flowers, as Helleborus, Epimedium, Viola, &c., and as an exceptional occurrence I have seen them in Ranunculus repens, while in Eranthis hyemalis transitions may frequently be seen between the flat outer segments of the perianth and the tubular petals. To Dr. Sankey, of Sandywell Park, I am indebted for the flower of a Pelargonium, in which one of the petals had the form of a cup supported on a long stalk. This cup-shaped organ was placed at the back of the flower, and had the dark colour proper to the petals in that situation. I have seen a petal of Clarkia similarly tubular, while some of the cultivated varieties of Primula sinensis exhibit tubular petals so perfect in shape as closely to resemble perfect corollas.
Fig. 9.—Eranthis hyemalis. Transition from flat sepal to tubular petal.
Like the petals, the stamens, and even the styles, assume a hollow tubular form. This change of form in the case of the stamens is, of course, usually attended by the petaloid expansion of the filament, or anther, and the more or less complete obliteration of the pollen sacs, as in Fuchsias, and in some double-flowered Antirrhinums.[20] So also in some semi-double varieties of Narcissus poeticus, and in Aquilegia. By the late Professor Charles Morren, this affection of the stamens and pistils was called Solenaidie,[21] but as a similar condition exists in other organs, it hardly seems worth while to adopt a special term for the phenomenon, as it presents itself in one set of organs.
In many of these cases it is difficult to say whether the cup-like or tubular form is due to a dilatation or hollowing out of the organ affected, or to a fusion of its edges. The arrangement of the veins will in some cases supply the clue, and in others the regularity of form will indicate the nature of the malformation, for in those instances where the cup is the result of expansion, its margin is more likely to be regular and even than in those where the hollow form is the result of fusion.
Cohesion of several organs by their margins:—leaves, &c.—The union of the margins of two or more different organs is of more common occurrence than the preceding, the leaves being frequently subjected to this change. Occasionally, the leaflets of a compound leaf have been observed united by their margins, as in the strawberry, the white trefoil, and others. Sometimes the union takes place by means of the stalks only. I have an instance of this in a Pelargonium, in Tropæolum majus, and Strelitzia regina; in other cases, the whole extent of the leaf becomes joined to its neighbour, the leaves thus becoming completely united by their edges, as in those of Justicia, oxyphylla.[22] M. Clos[23] has observed the same thing in the leaves of the lentil Ervum lens, conjoined with fasciation of the stem, and many other examples might be given. Some of the recorded cases are probably really due to fission of one leaf into two rather than to fusion. Although usually the lower portions of the leaf are united together, leaving the upper parts more or less detached, there are some instances in which the margins of the leaf at their upper portion have been noticed to be coherent, while their lower portions, with their stalks, were completely free.[24]
Cohesion of the leaves frequently accompanies the union of the branches and fasciation as might have been anticipated. Moquin cites the fenestrated leaves of Dracontium pertusum, as well as some cases of a similar kind that are occasionally met with, as instances of the cohesion of the margins at the base and apex of the leaf, which thus appears perforated. This appearance, however, is probably due to some other cause. When the leaves are verticillate and numerous, and they become coherent by their margins, they form a foliaceous tube around the stem. When there are but two opposite leaves, and these become united by their margins, we have a state of things precisely resembling that to which the term connate is applied.
Fusion of the edges of the cotyledons also occasionally takes place, as in Ebenus cretica.[25] It has also been observed in Tithonia, and is of constant occurrence in the seed leaves of some Mesembryanthema. This condition must be carefully distinguished from the very similar appearance produced by quite a different cause, viz., the splitting of one cotyledon into two, which gives rise to the appearance as if two were partially united together.