An important change, which marked the transition from cryptogams to flowering plants, must now be mentioned, and to this the animal kingdom furnishes a striking analogy. The lowest vertebrates, such as fishes, are oviparous; the ova are discharged and afterward incubated. Mammals, on the other hand, are viviparous; the young are hatched within the body of the parent. The young of the kangaroo and other marsupials, which constitute the lowest order of mammals, are still very immature at birth. Analagous conditions are found among plants. Cryptogams are all oviparous; the macrospore, which may be regarded as the ovum or egg, separates from the parent plant before fertilization. Phanerogams, on the other hand, may be described as viviparous, since they retain the macrospore or ovum until it has developed an embryo. The presence of an embryo constitutes the distinction between a seed and a spore. Unless an embryo be present a seed can not germinate, since germination is simply the emergence of the embryo from the coats of the seed. An extreme case of this retention is seen in the mangrove, where the seed germinates while still attached to the tree; the embryo sends down its long radicle into the mud, and only quits its hold of the parent when it has become firmly established. Orchids and many parasitic plants have seeds with exceedingly minute and imperfect embryos, recalling the undeveloped offspring of the marsupials. The retention of the egg is attended with a manifest advantage; plainly the viviparous method of reproduction, which obtains in the higher divisions of the two organic kingdoms, is much more economical than the other. By the change to the viviparous condition, several structures present in the cryptogams are rendered useless, and a disused organ invariably degenerates; the prothallus and its adjuncts, having no longer any function to perform, must inevitably begin to atrophy. The rudimentary structures appearing in the embryo-sac of phanerogams can in this way be accounted for. The life-history of a cryptogam extends, as we have seen, to two volumes; it now appears that the life-history of a phanerogam is a second edition, of the same story, somewhat abridged and completed in a single volume. The life-history of certain ferns occasionally undergoes a corresponding abbreviation. In the phenomena of apospory and apogamy we have departures from the ordinary course of development, closely akin to what would be required for the conversion of a cryptogam into a phanerogam. Apospory occurs when the production of spores is omitted, the prothallus growing immediately on the fern frond; apogamy, when the female organs are not developed, and the frond is formed by vegetative growth directly from the prothallus.