Fig. 38.—Flower of Œnanthe crocata, in which the five sepals were completely detached from the ovaries, here three in number and destitute of stylopods.
In Umbelliferæ, a detachment of the calyx from the ovaries frequently occurs, sometimes without any other change; at other times attended by more serious alterations. So far as can be judged from exceptional occurrences of this kind, it would appear that in this order the axis or flower-stalk does not, in any material degree, enter into the composition of the fruit.
In the Rubiaceous genus Bikkhia, as mentioned by Duchartre, the ovary is completely inferior, but when the fruit arrives at maturity four small leaves are detached from its surface which had previously adhered to it, and which it seems reasonable to consider as the sepals.
In Campanulaceæ a similar separation of calyx from the ovary may be occasionally met with. On the other hand, the occasional formation of a leaf on the inferior ovary of those plants would indicate the axial nature of the fruit. In Campanumæa and Cyclodon the calyx is inferior, while the corolla is superior. In the last-named genus this peculiarity "is carried to the highest degree, the sepals being, in C. parviflorum, placed on the peduncle of the flower far removed from the base of the corolla and ovary, whilst in C. truncatum and in Campanumæa they adhere to the base of the tube of the corolla."[87] In this order, then, as in Saxifragaceæ, Bruniaceæ, &c., no hasty conclusion should be drawn as to the nature of the fruit. In Brunia microphylla the ovary is superior, enclosed within but not adnate to the cup-like calyx, to which latter, however, the petals and stamens are attached.
In Onagraceæ (Jussieua), as also in Cactaceæ (Opuntia), buds have been observed on the surface and edges of the inferior ovary. Indeed, in the former genus, they have been produced artificially, but as buds may be formed on foliar as well as on axial organs, the fact cannot be made great use of in support either of the foliar or axial nature of the inferior ovary. In Epilobium, I have met with four perfect leaves at the summit of the ovary, in the place usually occupied by the sepals. This would also favour the notion that the axis entered into the constitution of the fruit in this genus.
Mr. B. Clarke, in his 'New Arrangement of Phanerogamous Plants,' p. 4, cites a case wherein the perianth was completely detached from the surface of the ovary in Cannabis sativa.
It must be borne in mind that some of the recorded instances of change in the relative position of the calyx and pistil ought more properly to be referred to a substitution of carpels for stamens, as in Begonia, Fuchsia, &c. Among Cucurbitaceæ, examples have been recorded, both of the detachment of the calyx from the ovary,[88] and of the partial conversion of some of the anthers of the male flower to carpels.
The very singular mode of germination of Sechium edule in which the fruit, instead of rotting, becomes thickened into a kind of rhizome or tuber, is a fact that should not be overlooked in investigating the true nature of the fruit in this order.
The following are the genera in which the change has been most frequently observed: