I] 



STRUCTURE OF THE SPOROPHYTE 



9 



are peripheral and two or sometimes more in number, the phloem-groups 

 alternating with them. In fact the root of a Fern is constructed essentially 

 on the plan of that in Seed-Plants, As there is no secondary thickening 

 the roots of Ferns are all fibrous. The lateral roots arise opposite to the 

 protoxylems, and there they originate from definite cells of the endodermis, 

 which may often be recognised beforehand by their size and contents. 



While we recognise the substantial similarity of Ferns and Seed-Plants 

 in respect of form and structure of stem, leaf, and root, these plants differ 

 in the construction of their apical merislems. In Seed-Plants these are 

 small-celled tissues, and more or less definitely stratified. In Ferns such as 

 Osmtinda, Dryopteris or Polypodiuni, a single large cell, the apical or iiiitial 

 cell, occupies the tip of each growing part. It has a definite shape, and 

 segments are cut off from its sides in definite succession. As the whole 



Fig. lo. Apex of stem of Osmunda lYgalis, seen 

 from above, showing the three-sided apical cells 

 of stem and of leaf shaded. The successive seg- 

 ments of the apical cell form the whole of the 

 apical cone. ( x 83.) 



Fig. II. Young leaf of Ceratopteris, in sur- 

 face view, after Kny ; showing two-sided 

 apical cell; and the marginal series, con- 

 tinuous round the young pinnae. The 

 latter do not correspond to the segments 

 from the apical cell. 



tissue of the stem, leaf, or root is derived from such segments, the whole of 

 each part is referable in origin to its apical cell, which maintains its identity 

 throughout. The form of the cell in roots, in most stems, and in some leaves 

 {Osmunda) is that of a three-sided pyramid; but where the organ is flat- 

 tened, as in some stems {Pteridiuvt) and almost all leaves, it has two convex 

 sides and is shaped like half of a biconvex lens. In the former case the 

 segments are cut off in regular succession from the three sides (Fig. 10), in 

 the latter alternately from the two sides (Fig. 11). The further subdivision 

 of the segments to form the tissues is represented in surface-view for the 

 case of Osimmda in Fig. 10 : and Fig. 1 1 shows in the surface-view of a 

 }^oung leaf of Ceratopteris how the whole member may be built up from 

 such segments. In roots the segmentation is complicated by the origin of 

 the root-cap. This is provided by a segment cut off from the frontal face 



