TRUE ferns were not nearly so common in the Coal Measure period as was formerly supposed, but their existence is well-attested by petrified remains ofand as well as spore-sacs. The fossils fall into two main sections : a group of ‘early ‘ferns which first appear in the Upper Devonian, with a fairly simple type of anatomical structure, and a group of tree-ferns which are characteristic of the later part of the era, with a very elaborate anatomy. The curious thing about the former is that many of them had most unusual fronds borne in definite rows, which branched repeatedly in planes at right angles to each other, so that they must have looked quite unlike any ferns we know at present, while the really fern-like fronds of the Carboniferous period were many of them not ferns at all. These early ferns seem to be another extinct specialised stock, though they show points of contact with certain later families of ferns, and especially with the family to which the Royal Fern belongs.
The Early Ferns may also be related to the tree-ferns of the later PaUeozoic, and although the latter, too, are extinct, it is commonly believed that they are allied to certain living tropical ferns of the family Marattiacea;. Their petrified stems are particularly common in certain Permian Beds of Saxony, and have long been familiar, when cut and polished, as decorative objects under the name of ‘starling-stones.’ They obtained the name because of the mottled appearance of the thick mass of feltedwhich grew round the stem. More is known about the stems than about the foliage and reproductive organs of most of these ferns; with the fertile fronds bearing spore-sacs, it is difficult to know whether or not they
are really true ferns or whether they are the pollen-bearing organs of plants such as those we are about to describe.