662 PROFESSOR F. O. BOWER ON 



The method of preparation has been a very simple one. The leaves are made 

 transparent by eau de Javelle ; stained with safranin ; cleared in bulk in oil of 

 cloves ; and mounted in Canada balsam. The result is a transparency, with the 

 vascular tissue so stained that the connections, even those at the base of the blade, 

 can be readily followed. 



The relation between branching of the leaf itself and branching of the veins is 

 obviously a close one. It suggests that where the branch-veins run parallel the 

 venation represents a latent branching of outline, and that originally distinct lobes 

 have been " webbed," or laterally merged together. This seems highly probable in 

 simple cases, but it may be doubted whether the branching of veins is always the 

 equivalent of a webbed branching of outline. In complex cases where reticulation 

 occurs there can be no question but that vascular strands have been formed de novo. 

 At the moment it is difficult to make any general statement on this point. It is best 

 to be fully aware of the difficulty, and to recognise provisionally that a close relation 

 subsists between branching and venation. When a large number of examples have 

 been considered in detail it may be possible to draw some more definite conclusion. 



Comparative study of the leaf-architecture of the Pteridophyta is the natural 

 foundation for a similar inquiry for the Spermophyta. The actual lines of evolu- 

 tion may be obscure, but there can be no doubt that the one will illuminate the 

 other. This was the view of Hofmeister, Sachs, Prantl, and other morphologists. 

 Accordingly we v may be prepared to trace at least analogies, if not some nearer degree 

 of resemblance, between the construction of the leaves of the more primitive 

 Pteridophyta and that of the relatively advanced Seed-bearing Plants. No special 

 observations relating to the latter will be described here, and it may be that further 

 inquiry will be necessary before full advantage can be taken of the basis for com- 

 parison which the Pteridophyta can supply. But at the close of this memoir 

 some attempt will be made to carry onwards to the Flowering Plants the applica- 

 tion of the results acquired, which may help to elucidate the architecture of their 

 •leaves also. 



Description of the Architecture of the Leaves of the Pteridophyta, with 

 special reference to their Primordial Leaves. 



Lycopodiales. 



In all of these the leaf is relatively small, and its construction is always simple 

 both in form and venation. It was, however, of considerable actual size in the 

 Lepidodendraceae and Sigillariacese, as it is also in the living Isoetes. It is traversed 

 by a single unbranched vascular strand. The only known exception to this among 

 living or fossil Lycopods is in certain Sigillarias, where in the middle region of the 

 leaf there are two equal strands.* These fuse towards the apex, and the whole 



* Scott, Studies in Fossil Botany, p. 230, fig. 95, of Sigillaria sulcata. 



