Double flowers of Orchidaceæ generally arise from petalification of the filaments, with or without other coincident changes. What makes double flowers in this order the more interesting is the development, in a petaloid condition, of some or all of those stamens which under ordinary circumstances are wholly suppressed, so that the morphological structure of the flower, at first a matter of theory, becomes actually realised. Fig. 156 is a diagram showing the presence of two additional labella within the ordinary one in a species of Catasetum, and representing two petaloid stamens, thus evidently completing the outer staminal whorl, of which there is usually but a single representative (see Peloria, Multiplication, Prolification). In some of these double orchids it is, however, necessary not to confound a petaloid condition of the existing column with the development of usually suppressed stamens in a petaloid form. Thus, in Lycaste Skinneri the column is frequently provided with two petal-like wings, which might readily be supposed to be two stamens of the inner whorl adherent to the column; a little attention, however, to the relative position of these adventitious wings is generally sufficient to enable the observer to ascertain the true nature of the appearance.[306]
Fig. 156.—Diagram of flower of Catasetum, with two labella.
Some forms of duplicate or hose in hose corollas are apparently due, not so much to the formation of a second corolla within the first, as to the presence of an inner series of petal-like stamens, which, by their cohesion, form a second pseudo-corolla within the first. The staminal nature of this pseudo-corolla is inferred from the occasional presence of anthers on it.[307] In Datura fastuosa, as well as in Gloxinia, a pseudo-corolla of this kind sometimes occurs with the addition of a series of petaloid stamens attached to its outer surface.[308]
When the petalody specially affects the anther-lobes, as in Arbutus, Petunia, Fuchsia, &c., the venation of the petal-like portion is very frequently laminar, thus tending to show that the anther is in such cases really a modification of the blade of the leaf; but as, on the other hand, we often find petal-like filaments bearing pollen-sacs on their sides, it is clear that we must not attribute the formation of pollen to the blade of the leaf only, but we must admit that it may be formed in the filament as well.[309]
Fig. 158.—Portion of a double columbine (Aquilegia), showing petalody of the connective.