PTEEIDOPHYTA— PILICIN^ 129 



leaves are sometimes of three kinds, the foliage leaves, some- 

 times entire, but usually much and repeatedly pinnate ; the 

 sporophylls, which in many ferns are exactly like the foliage 

 leaves in appearance, but. in others are much modified; and 

 scaly leaves, rarely found, and then only on subterranean 

 rhizomes. The vernation of the leaves is very markedly circin- 

 ate, the main axis and all its branches or pinnae emerging from 

 the earth rolled up with their lower surfaces outwards. The 

 stem and bases of the leaf stalks are furnished when young with 

 numerous scaly hairs, called ramenta, which are multicellular 

 and sometimes glandular. 



The growth in length of the stem is always brought about 

 by the divisions of a pyramidal apical cell of either two or three 

 sides, the apex of the pyramid being directed inwards. It soon 

 shows a differentiation into dermatogen, periblem, and plerome. 

 At first the stem is always monostelic, but this condition in 

 most cases soon gives way to polystely, which persists throughout 

 its length {fig. 884). The separate steles are usually gamodesmic, 

 the bundles of which they are composed being completely 

 united together, presenting the appearance of a central mass 

 of wood with two or three strands of protoxylem, almost sur- 

 rounded by bast, though not entirely, as the latter does not 

 wrap round the narrow end of the woody mass. The whole is 

 enclosed by a pericycle and an endodermis, the latter belonging 

 to the fundamental tissue {fig. 885). Sometimes a stele wiU 

 consist only of a single bundle {fig. 886), though generally fusion 

 takes place. The steles as viewed in longitudinal section of the 

 stem, or better, as isolated by maceration, are found to anasto- 

 mose together very irregularly, forming a meshwork, from the 

 angles of which branches go off to enter the leaves. The nature 

 of the anastomosis is largely determined by the number and 

 size of the leaves. 



The bundles are said to be concentric. As they are usually 

 placed two or three together in the stele and fused laterally, this 

 is not very apparent, the whole stele seeming rather to deserve 

 this name, the fused bast masses surrounding the fused wood 

 masses. A single bundle, however, when found free {fig. 886), is 

 seen to be concentric ; and a stele is usually composed of two or 

 more placed so close together that the bast is not developed on 

 their contiguous faces. The lateral fusion of bast and wood of the 

 original bundles thus gives rise to the mass of wood in the stele 

 with its peripheral envelope of bast. The steles should perhaps 

 be called bi-coUateral rather'than concentric, as the bast is not 



VOL. II. K 



