There are, however, some difficulties to face in attempting to decipher the history of the Filicineae as recorded in the earth’s crust. Few fossil plants are so familiar as the well-preserved carbonaceous impressions of compound leaves on the shales of our Coal-Measures, which were referred by older authors to recent genera and species of ferns and accepted by later writers as undoubted examples of Palaeozoic ferns. The common belief in the dominance of ferns in Palaeozoic floras is reflected in the novelist’s description of the Carboniferous period, “when the forms of plants were few and often of the fern kind[678].” We now know that very many of these Carboniferous leaves belonged to plants differing widely in morphological characters from the modern genera to which they exhibit so deceptive a resemblance. These pseudo-ferns, recently christened Pteridosperms or seed-bearing fern-like plants, are dealt with in a later chapter. The discovery of this extinct group has added enormously to our knowledge of plant-evolution and at the same time has rendered much more difficult the task of unravelling the past history of the true ferns. As soon as it was demonstrated that many familiar Palaeozoic “ferns” are not ferns, some authors went far towards concluding that however close might be the agreement between fossil and recent leaves suspicion of close relationship must be set aside. Like the earlier writers who described fossils as _lusus naturae_ fashioned by devilish agency to deceive too credulous man, the discovery of seed-bearing plants with the foliage of ferns threatened to disturb the mental balance of palaeobotanists. The fact is, we cannot in some cases determine from leaf-form alone whether or not a fossil is a true fern; we may, as Professor Bower[679] suggests, regard all fern-like fossils as ferns until they are proved to be Pteridosperms, or in a spirit of scientific scepticism, we may at once admit that many Palaeozoic fern-like leaves must await further evidence before their true position can be determined. It is impossible, as Zeiller[680] says, in the present state of our knowledge to range fern-like Palaeozoic plants in two groups, one referred to Filicineae and the other to the Pteridosperms. The following classification of the Filicales is based on that adopted by Prof. Engler in the latest edition of his _Syllabus_[681] and on the results of Bower’s[682] excellent work on the spore-bearing members of recent ferns.