We have already seen that any animal devotes to reproduction the balance between income and expenditure of nutriment. Now, the digestive system is here well developed, and the income is large. But we have already noticed that, as animals grow larger, the ratio between the digestive surface and the mass to be supported grows continually smaller. On account of size alone the mammal has but a small balance. But the amount of expenditure is proportional to the mass and activity of the muscular and nervous systems. And the mammal is, and from the beginning had to be, an exceedingly active, energetic, and nervous animal. The income has increased, but the expenses have far outrun the increase. The mammal can devote but little to reproduction.
Moreover, it requires a large amount of material to form a mammalian egg, such as that of the monotreme. It requires indefinitely more nutriment to build a mammal than a worm, for the former is not only larger and more perfect at birth; it is also vastly more complicated. The embryonic journey has, so to speak, lengthened out immensely. One monotreme egg represents more economy and saving than a thousand eggs of a worm. Moreover, where the individuals are longer lived and the generations follow one another at longer intervals, the number of favorable variations and the possibility of conformity to environment through these is greatly lessened. In such a group it is of the utmost importance that every egg should develop; the destruction of a single one is a real and important loss to the species. It is not enough to produce such an egg; it must be most scrupulously guarded. Even the egg of the platypus is deposited in a nest in a hole in the bank, and the female Echidna carries the egg in a marsupial pouch until it develops.
Notice further that among certain species of fish, amphibia, and reptiles, the females carry the eggs in the body until the embryos or young are fairly developed. Viviparous forms are unknown by birds, probably because this mode of development is incompatible with flight, their dominant characteristic. Putting these facts together, what more probable than that certain primitive egg-laying mammals should have carried the eggs as long as possible in the uterus. The embryo under these conditions would be better nourished by a secretion of the uterine glands than by a very large amount of yolk. The yolk would diminish and the egg decrease in size, and thus the marsupial mode of development would have resulted. And, given the marsupial mode of development and an embryo possessing an allantois, it is almost a physiological necessity that in some forms at least a placenta should develop. That the placenta has resulted from some such process of evolution is proven by its different stages of development in different orders of mammals. And even the feeblest attachment of the allantois of the embryo to the wall of the uterus would be of the greatest advantage to the species.