The amphibia and reptiles also pass into their modern types, after a period of generous expansion. Primitive frogs and toads make their first appearance in the Tertiary, and the remains are found in European beds of four-foot-long salamanders. More than fifty species of Tertiary turtles are known, and many of them were of enormous size. One carapace that has been found in a Tertiary bed measures twelve feet in length, eight feet in width, and seven feet in height to the top of the back. The living turtle must have been nearly twenty feet long. Marine reptiles, of a snake-like structure, ran to fifteen feet in length. Crocodiles and alligators swarmed in the rivers of Europe until the chilly Pliocene bade them depart to Africa.

In a word, it was the seven years of plenty for the whole living world, and the expansive development gave birth to the modern types, which were to be selected from the crowd in the subsequent seven years of famine. We must be content to follow the evolution of the higher types of organisms. I will therefore first describe the advance of the Tertiary vegetation, the luxuriance of which was the first condition of the great expansion of animal life; then we will glance at the grand army of the insects which followed the development of the flowers, and at the accompanying expansion and ramification of the birds. The long and interesting story of the mammals must be told in a separate chapter, and a further chapter must be devoted to the appearance of the human species.

We saw that the Angiosperms, or flowering plants, appeared at the beginning of the Cretaceous period, and were richly developed before the Tertiary Era opened. We saw also that their precise origin is unknown. They suddenly invade a part of North America where there were conditions for preserving some traces of them, but we have as yet no remains of their early forms or clue to their place of development. We may conjecture that their ancestors had been living in some elevated inland region during the warmth of the Jurassic period.

As it is now known that many of the cycad-like Mesozoic plants bore flowers—as the modern botanist scarcely hesitates to call them—the gap between the Gymnosperms and Angiosperms is very much lessened. There are, however, structural differences which forbid us to regard any of these flowering cycads, which we have yet found, as the ancestors of the Angiosperms. The most reasonable view seems to be that a small and local branch of these primitive flowering plants was evolved, like the rest, in the stress of the Permian-Triassic cold; that, instead of descending to the warm moist levels with the rest at the end of the Triassic, and developing the definite characters of the cycad, it remained on the higher and cooler land; and that the rise of land at the end of the Jurassic period stimulated the development of its Angiosperm features, enlarged the area in which it was especially fitted to thrive, and so permitted it to spread and suddenly break into the geological record as a fully developed Angiosperm.

As the cycads shrank in the Cretaceous period, the Angiosperms deployed with great rapidity, and, spreading at various levels and in different kinds of soils and climates, branched into hundreds of different types. We saw that the oak, beech, elm, maple, palm, grass, etc., were well developed before the end of the Cretaceous period. The botanist divides the Angiosperms into two leading groups, the Monocotyledons (palms, grasses, lilies, orchises, irises, etc.) and Dicotyledons (the vast majority), and it is now generally believed that the former were developed from an early and primitive branch of the latter. But it is impossible to retrace the lines of development of the innumerable types of Angiosperms. The geologist has mainly to rely on a few stray leaves that were swept into the lakes and preserved in the mud, and the evidence they afford is far too slender for the construction of genealogical trees. The student of living plants can go a little further in discovering relationships, and, when we find him tracing such apparently remote plants as the apple and the strawberry to a common ancestor with the rose, we foresee interesting possibilities on the botanical side. But the evolution of the Angiosperms is a recent and immature study, and we will be content with a few reflections on the struggle of the various types of trees in the changing conditions of the Tertiary, the development of the grasses, and the evolution of the flower. In other words, we will be content to ask how the modern landscape obtained its general vegetal features.

Broadly speaking, the vegetation of the first part of the Tertiary Era was a mixture of sub-tropical and temperate forms, a confused mass of Ferns, Conifers, Ginkgoales, Monocotyledons, and Dicotyledons. Here is a casual list of plants that then grew in the latitude of London and Paris: the palm, magnolia, myrtle, Banksia, vine, fig, aralea, sequoia, eucalyptus, cinnamon tree, cactus, agave, tulip tree, apple, plum, bamboo, almond, plane, maple, willow, oak, evergreen oak, laurel, beech, cedar, etc. The landscape must have been extraordinarily varied and beautiful and rich. To one botanist it suggests Malaysia, to another India, to another Australia.

It is really the last gathering of the plants, before the great dispersion. Then the cold creeps slowly down from the Arctic regions, and begins to reduce the variety. We can clearly trace its gradual advance. In the Carboniferous and Jurassic the vegetation of the Arctic regions had been the same as that of England; in the Eocene palms can flourish in England, but not further north; in the Pliocene the palms and bamboos and semi-tropical species are driven out of Europe; in the Pleistocene the ice-sheet advances to the valleys of the Thames and the Danube (and proportionately in the United States), every warmth-loving species is annihilated, and our grasses, oaks, beeches, elms, apples, plums, etc., linger on the green southern fringe of the Continent, and in a few uncovered regions, ready to spread north once more as the ice creeps back towards the Alps or the Arctic circle. Thus, in few words, did Europe and North America come to have the vegetation we find in them to-day.

The next broad characteristic of our landscape is the spreading carpet of grass. The interest of the evolution of the grasses will be seen later, when we shall find the evolution of the horse, for instance, following very closely upon it. So striking, indeed, is the connection between the advance of the grasses and the advance of the mammals that Dr. Russel Wallace has recently claimed ("The World of Life," 1910) that there is a clear purposive arrangement in the whole chain of developments which leads to the appearance of the grasses. He says that "the very puzzling facts" of the immense reptilian development in the Mesozoic can only be understood on the supposition that they were evolved "to keep down the coarser vegetation, to supply animal food for the larger Carnivora, and thus give time for higher forms to obtain a secure foothold and a sufficient amount of varied form and structure" (p. 284).

Every insistence on the close connection of the different strands in the web of life is welcome, but Dr. Wallace does not seem to have learned the facts accurately. There is nothing "puzzling" about the Mesozoic reptilian development; the depression of the land, the moist warmth, and the luscious vegetation of the later Triassic and the Jurassic amply explain it. Again, the only carnivores to whom they seem to have supplied food were reptiles of their own race. Nor can the feeding of the herbivorous reptiles be connected with the rise of the Angiosperms. We do not find the flowering plants developing anywhere in those vast regions where the great reptiles abounded; they invade them from some single unknown region, and mingle with the pines and ginkgoes, while the cyeads alone are destroyed.

The grasses, in particular, do not appear until the Cretaceous, and do not show much development until the mid-Tertiary; and their development seems to be chiefly connected with physical conditions. The meandering rivers and broad lakes of the mid-Tertiary would have their fringes of grass and sedge, and, as the lakes dried up in the vicissitudes of climate, large areas of grass would be left on their sites. To these primitive prairies the mammal (not reptile) herbivores would be attracted, with important results. The consequences to the animals we will consider presently. The effect on the grasses may be well understood on the lines so usefully indicated in Dr. Wallace's book. The incessant cropping, age after age, would check the growth of the larger and coarser grasses give opportunity to the smaller and finer, and lead in time to the development of the grassy plains of the modern world. Thus one more familiar feature was added to the landscape in the Tertiary Era.