The Phylogenetic Classification of Flowering Plants

Abstract

THOUGH “the abominable mystery,” as Darwin called the problem of the origin of Flowering Plants, is by no means solved, there has been before the botanical world for some years a theory of their origin which is consonant with the derivation of all forms of existing flowers from the Ranalean type. Through Wieland's brilliant elucidation of the structure of the fossil Bennettitean fructification at the beginning of this century, botanists became acquainted for the first time with an unexpected bisexual seed-bearing cone, in which the two kinds of sporophylls bore the same relative position to one another on the axis as they (the stamens and carpels) invariably do in the angio-spermous hermaphrodite flower; and further, such a cone was subtended by a number of bracts resembling a perianth. The temptation naturally was great to suggest a real bond of affinity between the Flowering Plants and these Cycadean-like Mesozoic plants, the Bennettitales (Cycadeoideas). A theory was worked out to this effect. Though the peculiar nature of the female part of the cone precludes the direct origin of the Angiosperms from the Bennettitales, the view was put forward that the two groups had diverged from common ancestors with a generalised type of flower-like cone-the anthostrobilus, as it was called. These ancestors, it was thought, probably arose from the seed-ferns (Pteridosperms). It was further conjectured that the Angiosperms owed their being to the substitution of insect-pollination (entomophily) for wind-pollination (anemophily), and it has even been hazarded later that this type of cone, the anthostrobilus, may have been evolved in response to insect-visitation. The theory also provided a resting-place for that small puzzling group of Gymnosperms, the Gnetales, and accounts for the peculiar male (morphologically hermaphrodite) “flower” of Welwitschia- the stumbling - block to those botanists who endeavour to derive the Gnetales from the Conifers, a group characterised as a whole by possessing unisexual cones. The whole speculation, which has rightly or wrongly been termed the “strobilus theory of angiospermous descent,” still awaits confirmation or refutation. It has with some botanists lost favour on general grounds. The recent discoveries of palaobotany show that many of the main groups of vascular land plants can be traced right back to Devonian times as independent 'lines, and Dr. A. H. Church in a recent stimulating memoir has speculated as to the possibility of these main phyla having been even differentiated in the sea independently from one another. Thus it is just possible that the Angiosperms may represent such an independent line, originating from a distinct group of Algae. Their resemblance then to the Bennettitales would become merely an interesting parallelism. This then may be said to be the chief criticism that can as yet be levelled against the strobilus theory. Dr. Scott, in his latest edition of his “Studies in Fossil Botany,” sums up in favour of the theory.2 If the Angiosperms throughout the ages have been independent of any other vascular group, one wonders why obvious traces of them have never come to light in the Palaeozoic rocks. It is difficult to believe that they can, as it were, have stepped out of the sea, fully differentiated, in late Mesozoic times. One of the astonishing facts of palaeobotany has been the sudden appearance of Flowering Plants in late Cretaceous times. Dicotyledonous stems are now known, however, from lower Cretaceous rocks, and recently Mr. Hamshaw Thomas has brought to light some remarkable fossil fruits (the Caytoniales), resembling those of Angiosperms, from the middle Jurassic. Possibly then ere long some fresh light may be thrown on the problem.