﻿MR. JOHN SCOTT ON THE TREE FERNS OF BRITISH SIKKIM. 29 



normal thickness has been attained, all further peripheral increase ceases, and the 

 axis grows upward in a cylindrical form. In Palms the growing point is conical, and 

 in tree ferns terminally flattened ; so that in the one case the nascent vascular hundles are 

 almost horizontally arranged on the apex, in the other they are, from the beo'innin 



s —- ^^^o 



vertical ; but in both all cellular increase of the body of the axis has ceased ere the fronds 

 have attained their full development. Another point in which they agree, is in the 

 development of a system of free vascular bundles, which, originating in the apex of the 

 stem, grow upwards into the fronds and downwards in the axis. In other respects they 

 present many important points of difference, as in the free anastomoses of the fibro- 

 vascular bundles, and the formation of a woody circle surrounding the central cellular 

 axis, which, however, is partially represented in certain Palms {vide Plate XII.), as JEJu- 

 terj)e oleracea, by a dense zone of vascular bundles, which very distinctly separates the 

 central, in which cellular tissue predominates, from the pcripherical and essentially 

 vascular. The tissues of this median zone are developed, I believe, exclusively from 

 below upwards, the growing points surrounding the apex of the stem, and passing out- 

 wards into the fronds. Dense though this zone is, the bundles do not, in any case that I 

 have observed, inosculate with each other. Another important difference is presented in 

 the passage of these bundles to the fronds. In Perns small ramifications of the bundles 

 only pass into any given frond, while in Palms the entire bundle enters the frond. I 

 need scarcely remark that there are also important differences in the minute structure of 

 woody bundles of Perns and Palms, the former being much more simple, though 

 apparently sunilarly limited in their period of growth. In thus noting the structural 

 affinities and differences of the stems of tree ferns and palms, I by no means ascribe to 

 them an equal grade of organization ; tree ferns are decidedly inferior in rank. They 

 have, however, sufficient in common to justify the opinion of their being differently 

 diverged and progressed forms of a common ancestor, which must have been of great 

 antiquity, considering that both apparently presented much the same characteristics 



L 



in the Carboniferous epoch as they do now. 



The anatomical relations of the caudex of tree ferns to the less perfectly developed 

 dicotyledonous stem is of much interest in theoretical natural science, as showing us the 

 graduated steps by which the simpler condition of the fibre- vascular system and stem- 

 structure of the arborescent Filices is related (and that mayhap ancestorially) to that of 

 the more highly developed dicotyledon. The vascular system has been well distin- 

 guished by Schleiden as " simultaneous," « definite," and " indefinite." The first two 

 modes of developement are almost peciiliar to the higher cryptogams and monocoty- 

 ledons, though the first is the more especial characteristic of the former, as the second is 

 of the latter class, while the indefinite mode characterizes nearly the whole of the dicoty- 

 ledonous division. In the present remarks it is necessary to notice those only of the 

 latter series presenting the least-perfect forms. This, I believe, is represented by the 

 Nymphseas, which present the following anomalous arrangement — a central parenchy- 

 matous mass, interspersed with variously directed isolated woody bundles, with an hre- 

 gular circle of larger bundles surrpunding it. Prom the latter branches proceed through 

 the external parenchymatous mass towards the appendicular organs, and present a vcrv 



