322 FOSSIL PLANTS OF THE COAL-MEASURES. admirable drawings in plate xvi. fig. 4. The nature of the secondary tissues has been satisfactorily determined from an examination of numerous roots in different stages of development. On the whole, the secondary wood is identical with that of the Calamite stem ; immediately opposite to each protoxylera group there is usually found a medullary ray, which in some cases may be traced for a long distance through the wood. The presence of this medullary ray affords another piece of confirmatory evidence as to the practical identity in structure between the roots of Calamites and those of recent plants. Of the phloem, fewer and less distinct traces have been detected. Passing to the cortical tissues, a definite structural feature is recognised in the three well-marked zones ; an inner cortex of continuous parenchyma, a middle lacunar cortex, and an outer zone of continuous parenchymatous tissue. In some of the smaller root-branches, an extremely interesting feature is pointed out in the innermost layers of the cortex; next to the primary phloem there occur two layers of cells corresponding in position to the endodermis and pericycle, but the cells of the two layers "fit on to each other as if they had had a common origin." This suggests the double endodermis characteristic of Equisetumy and, if established, affords another illustration of the very intimate connection between the arborescent equisetaceous plants of the Coal-Measures and the recent Horsetails. In the so-called rootlets of Calamites the pith of the older roots is usually absent ; and the structure may be diarch or tetrach. In the thicker roots as many as twenty-five groups of primary xylem and phloem occur. The outermost cortical layer is characterised by thick external walls, and is spoken of as the epidermoidal layer. It would seem that a layer of more delicate cells was originally ex- ternal to this protective epidermoidal tissue, but was cast off at a comparatively early stage of development. The genus Calamites affords one of the most striking examples among palaeozoic plants, of the very great importance of the application of detailed histological methods in palaeobotanical investigations. The memoir on Lyginodendron and Heterangiiim is undoubtedly one of the most important contributions to fossil botany published in recent years. The interest attached to the two genera is one which every student of plant evolution cannot fail to appreciate. None of the records of the past are of such importance as those which enable us to bridge over gaps between existing classes of organisms. We have previously learnt something of intermediate plant types, e. g. the palaeozoic genera Poroxylon, Protojntys, Mye- loxylon, and others ; but of Lyginodendron and Heterangiiim our knowledge has been rendered much more complete, and many new characters of primary importance are clearly demonstrated in this admirable piece of investigation. Lyginodendron. — Fragments of stems, petioles, and foliage of this genus are fairly abundant in the calcareous nodules or coal-balls of Lancashire and Yorkshire. Although no undoubted examples of impressions or casts have so far been recognised, it is possible to restore with a considerable degree of probability the general form
