On the other hand, it is not easy to admit such repeated risings and sinkings of surface as would be required, on this hypothesis, to form a series of coal strata. Perhaps we may most safely rest at present with the supposition that coal has been formed under both classes of circumstances, though in the latter only as an exception to the former. Upwards of three hundred species of plants have been ascertained to exist in the coal formation; but it is not necessary to suppose that the whole contained in that system are now, or ever will be distinguished. Experiments shew that some great classes of plants become decomposed in water in a much less space of time than others, and it is remarkable that those which decompose soonest, are of the classes found most rare, or not at all, in the coal strata. It is consequently to be inferred that there may have been grasses and mosses at this era, and many species of trees, the remains of which had lost all trace of organic form before their substance sunk into the mass of which coal was formed. In speaking, therefore, of the vegetation of this period, we must bear in mind that it may have comprehended forms of which we have no memorial. Supposing, nevertheless, that, in the main, the ascertained vegetation of the coal system is that which grew at the time of its formation, it is interesting to find that the terrestrial botany of our globe begins with classes of comparatively simple forms and structure. In the ranks of the vegetable kingdom, the lowest place is taken by plants of cellular tissue, and which have no flowers, (_cryptogamia_,) as lichens, mosses, fungi, ferns, sea-weeds. Above these stand plants of vascular tissue, and bearing flowers, in which again there are two great subdivisions; first, plants having one seed-lobe, (_monocotyledons_,) and in which the new matter is added within, (_endogenous_,) of which the cane and palm are examples; second, plants having two seed-lobes, (_dicotyledons_,) and in which the new matter is added on the outside under the bark, (_exogenous_,) of which the pine, elm, oak, and most of the British forest-trees are examples; these subdivisions also ranking in the order in which they are here stated. Now it is clear that a predominance of these forms in succession marked the successive epochs developed by fossil geology; the simple abounding first, and the complex afterwards. Two-thirds of the plants of the carboniferous era are of the cellular or cryptogamic kind, a proportion which would probably be much increased if we knew the whole Flora of that era. The ascertained dicotyledons, or higher-class plants, are comparatively few in this formation; but it will be found that they constantly increased as the globe grew older.