rogenous initial to the mature pollen grain, and may involve only the sporogenous tissue and its products, or the entire anther ; (b) in the ovary at any stage from the functioning megaspore to the maturing fruit, and may involve only the embryo-sac, or both embryo-sac and ovule, or the entire ovary; (c) in entire inflorescences. The degeneration is accompanied by, and probably caused by, the formation of a definite abscission layer across the peduncle, whereby the floral organs are cut off from nutriment, and is so common that functionally perfect flowers are rare. As a result of such degenerations four types of mature flowers are produced (a) physio- logically staminate (the ovary being functionless) in which the pollen may or may not be functional ; (b) physiologically ovulate, the stamens having been completely eliminated by degeneration ; (c) bisporangiate, having both stamens and ovary functional (very rare); (d) completely sterile, having functionless ovary and stamens completely eliminated. The author discusses shortly the bearing of these facts on the origin of dicliny, which he is inclined to consider may be due in other species also to gradual degenerations, and not as Strasburgher supposed to excessive mutation, nor to an absence of one sex in the seed, as suggested by Hoffmann, nor to hybridisation as Jeffrey believed.
Seedling Structure.
Compton, R. H. An investigation of the Seeding Structure in the Leguminosese. Jour. Linn. Soc. {Botany XLI No. 279, pp. 1 — 129 {June 1919.)
A description with illustrative figures is given of the distribution of the xylem and phloem in seedings of species taken from all sections of the order. It would be impossible in a brief note such as this to summarise or even indicate the facts and ideas which the author gives; but it is clear that there is considerable variation in the amount and arrangement of the xylem elements in the root from a solid tetrarc core, to a ring of eight separate Y-shaped masses (as seen in section.) Those interested would do well to read the paper. But two conclusions may be given here—one that the size of the seeds is correlated with the habit of the plant, trees having much heavier seeds than herbs or shrubs ; the other that as with Angiosperues in general the tree habit is primitive, the herbaceous derived. As to whether the epigeal or the hypogeal mode of germination is the more primitive it is impossible to say. Both occur sometimes in the same genus, and the change from one to the other has probably occurred repeatedly in the evolution of the order.
Pollination Mechanisms.
Wyley, R. B. The Pollination of Vallisneria spiralis. Bot. Gaz. I XIII, pp. 135—145.
The mechanism of pollination in the American plant differs markedly from that described by Koerner, and widely copied from his Pflamerileben. The chief points are that the staminate flowers float on their recurved sepals, till
