Leaving the threshold of the Mesozoic age we may glance for a moment at the records of plant-life preserved in the rocks classed by geologists as Cainozoic or Tertiary, the rocks which chronicle the events between the Mesozoic era and the relatively short period which began with the great Ice age and shades imper ceptibly into the historical age. The vegetation of the Cainozoic era was practically identical in its general composition with that of tropical and sub-tropical lands at the present day. Then as now flowering plants (Angiosperms) were the ruling dynasty and the majority of gymnosperms, ferns and other groups were essentially similar to their descendants which flourish at the present day.
The Palaeozoic forests were archaic and unfamiliar; those of the Cainozoic periods were definitely modern. This transformation was effected during the Mesozoic era; it is therefore in hopeful expectation that the student of ancient floras turns to the frag mentary samples of the vegetable world preserved in the rocks of the intervening epoch of geological history. From a study of Mesozoic floras we might expect to be able to connect the portions of the chain of life reconstructed from Cainozoic fossils with the much more ancient pieces disinterred from Palaeozoic strata. While it is true to say that palaeobotanical research has contrib uted much towards a knowledge of the successive floras of the Mesozoic era, it is equally true to say that we are still groping af ter clues which may eventually enable us to visualize the sequence of events in the course of the middle or critical stage in the history of the plant-world. The chief purpose of this article is to present in as concise a form as possible the evolutionary tendencies in the course of the long ages separating the Palaeozoic from the Caino zoic era, so far at least as they can be discovered from the meagre documents at our disposal. The imperfection of the geological record is well put by Sir Joseph Hooker : "We have not in a fossil ized condition a fraction of the plants that have existed, and not a fraction of those we have are recognizable specifically." So far no reference has been made to the Palaeozoic vegetation in the southern hemisphere. In order to obtain as complete a pic ture as possible of the immediate antecedents of the earliest Meso zoic floras it is necessary to include in our summary the greater part of the earth's surface on which we have any information. Owing to the difficulty of defining the boundary in some regions between the Carboniferous and Permian plant-beds it has been a common practice to employ the term Permo-Carboniferous as an admission of incomplete knowledge. We have already given a brief account of some of the salient features of the Permo-Car boniferous vegetation in the northern hemisphere. It is generally agreed that on the northern continents at the close of the Carbon iferous period the climatic conditions were genial and favourable to the development of a luxuriant vegetation, and that in the course of the Permian period folding of the earth's crust produced changes in the physical setting which reacted disastrously on the plant-world. We cannot separate the organic from the inorganic world; evolution must be considered from a double aspect, the evolution of the plants in relation to a changing environment.
The Glossopteris Flora.—A precise correlation of beds in widely separated parts of the world is by no means easy ; we cannot, for example, assert with confidence whether or not a flora revealed by a study of late Palaeozoic rocks in India and the southern hemisphere was actually contemporaneous with the f or ests of the northern hemisphere Coal age, or whether certain climatic conditions demonstrated by geologists in the lands south of the Equator synchronized with a strongly contrasted climate in the North. There is, however, a strong body of evidence in support
of the view that before the close of the Carboniferous period, when large areas on the northern continents were covered with forests, the southern lands were in the grip of an Ice age and sup ported a sparse vegetation markedly different from that found on the northern swamps. In the early days of the Carboniferous period closely allied or even specifically identical plants grew in the Arctic regions, in Australia and in many other widely sepa rated regions; the vegetation seems to have been remarkably uni form. In the later stages of the period, on the other hand, there was a well marked differentiation into two or more botanical prov inces occupied by more or less sharply contrasted floras. In the latter part of the Palaeozoic era there were two large con tinents in the northern hemisphere, one including North America, Greenland and Europe which is sometimes spoken of as Eria ; and another embracing a large part of Siberia known as Angara Land. Beyond the world-encircling Tethys sea, which washed the south ern shores of these great continents, was the vast continent of Gondwanaland, formed of what are now India, Australia, Africa and South America. In geological maps Gondwanaland is usually shown as a continuous land mass parts of which are asssumed to have foundered. If we adopted the hypothesis of Wegener, Gondwanaland would have to be represented as a portion of the crust formed of the present southern continents and India fitted into one another like pieces of a jig-saw puzzle. From this com pact mass, disrupted in the Mesozoic era by deep fissures, huge blocks slowly drifted away, like icebergs from a glacier, until they reached their present positions. The Permo-Carboniferous plants and sedimentary rocks of Gondwanaland indicate a state of affairs in the organic and inorganic worlds very different from that re vealed by the records in the northern hemisphere. In New South Wales, for example, rocks containing remains of Lower Carbonif erous plants agreeing closely with northern species are succeeded by thick masses of old boulder clays (tillites) which afford con vincing evidence of a glacial period at the close of the Palaeozoic era, a reign of ice probably longer and more widespread than that which in comparatively recent times held sway over North Amer ica and Europe. Geological investigations in Australia have estab lished the fact that before the end of the Carboniferous period the earth's crust was uplifted into mountain-ranges and conditions were produced favourable to the accumulation of ice and snow. Throughout geological history there were recurrent cycles of mountain-building and shifting of the scenes which produced new sets of factors conditioning plant-life; revolutions in the inorganic world caused transformations in the organic world. The story told by the series of plant-beds and glacial deposits in Australia, though differing in detail from that derived from a study of the corre sponding strata in India, South America and Africa is broadly speaking the same for Gondwanaland as a whole. Beds of boulder clay containing innumerable erratic blocks, of ten many tons in weight and resting on platforms grooved and striated by slowly moving rock-studded masses of ice, occur over thousands of square miles from one end of Gondwanaland to the other, from the Falk land Islands to Tasmania, from northern Australia to Afghanistan. The approximate areas which are occupied by glacial beds are in dicated in the map. The discovery of plant remains in the Argentine and in South Africa at the base of the glacial beds shows that in these regions, as probably elsewhere, the climatic conditions were not fatal to the existence of vegetation.