Study of the vascular supply to the carpels in the follicle bearing ranunculaceae

1. There is variation with regard to the number of bundles supplying each perianth segment, and the number of gaps formed. There is a tendency to reduce the number of gaps, and the number of traces entering the base of each, as we pass from the type with a large base of insertion, as in Trollius, to a type with a marrow base, as in Cimicifuga. The marginal bundles frequently branch on their outward passage, and also, occasionally, the midrib bundle. Apart from this branching in exceptional cases, the supply, fundamentally , resembles that of a leaf.2. The honey leaves and stamens each receive a single strand.3. In the carpellary supply, there is variation with regard to:- a. the number of bundles entering the base of a carpel. b. the number of receptacular bundles from which these traces are derived. c. the number of carpels each receptacular trace may supply.4. There is no correlation between the number of bundles in the receptacle, above the level of departure of the stamen traces and the number of carpels developed above this level.5. As a direct result of this, the degree to which these stelar bundles are divided, and the number of carpels developed which they supply, must depend upon the number of receptacular bundles, and the number of carpels developed. Thus, a comparatively small number of bundles must divide to a greater extent to supply a given number of carpels, than if a larger number of bundles were present, and vice versa.6. The whole of the stelar vascular tissue may or may not be used up in the formation of the ca:opellary traces, Both conditions may be found to occur in the one species - e.g. Caltha.7. There may be three or five traces enter the base of each carpel e.g. in Caltha, Trollius, and. Eranthis, where the midrib bundle divides in the pedicel, which is considered as an extension of the receptacle.8. The secondary vascular supply within the carpellary wall varies within fairly wide limits for each species.9. The main supply to the carpellary wall is, in the majority of species, derived from the marginal bundles, which give off branches. The terminal endings distant from the marginal bundle, either fuse directly with the midrib or end blindly in the carpellary wall. This is constant within a species.10. Branches may or may not arise from the midrib bundle within the carpel itself - e.g. Caltha and Cimicifuga davurica. Where present there is a tendency for these bundles to arise towards the carpel base, and pass out into the wall for varying distances. Normally they are not well developed. In Trollius, two branches constantly are given off, which continue parallel to the midrib. Again they tend to arise near the base of the carpel.11. These facts may be taken as support of the view that the primitive carpel may have been a palmately veined structure.12. The stylar supply, in the species where the number of parts in the gynoecium tends to be constant, is, of itself, constant. In Trollius and Caltha there is variation , the midrib, or marginal bundles, or both may supply the style. This variation also occurs with one gynoeciun.13. The fundamental vascular supply to a carpel is to that of a leaf. The venation within the carpellary wall, however, does not resemble that found in a true leaf.14. These results do not lend support to the theory of carpel polymorphism, not do they justify Thomas' interpretation of the probable origin of the carpel in such a type as Caltha.15. The present state of our knowledge of ancient Angiosperms, and of the more primitive types in our present day flora, only suffìce to show the tremendous gap in our knowledge of the probable primitive Angiosperm from which our present day carpel arose