M. \RATTIACEAE 



525 



It will be unnecessary to describe it in detail here : our object will be 

 rather to bring it into relation with the less complex systems of other 

 Ferns, and with the cognate fossils. This is most readily done by reference 

 to the seedling-structure, and to those genera which are less complicated 

 in their mature state ; for there is considerable variety of complexity in 

 the different living genera of the family. It is found that Kanlfi/ssia and 

 Archangiopteris are relatively simple, while Angioptcris is the most 

 complicated of all, Marattia and Danaea 

 taking a middle position. 



In the seedlings of them all the axis 

 is traversed by a monostele : in Danaea 

 simpliiifolia it has a solid xylem-core, 

 which is maintained till several leaf-traces 

 have been given off from it, naturally 

 without any leaf-gap : : it then becomes 

 crescentic, and expands into a dictyostele 

 with leaf-gaps, while a central strand or 

 commissure arises from the concavity of the 

 dictyostele, and pursues an upward course 

 with occasional fusions at the successive 

 leaf-gaps.- The same type of structure is 

 closely followed in the mature stem of 

 Archangiopteris? though on a simpler scale ; 

 in fact, this stem still retains at maturity a 

 stage rapidly passed through by the young 

 plants of other more complex genera. A 

 similar vascular system, consisting of a 

 cylindrical dictyostele, with normally a 

 single central strand, is found also in the pi^f'shf/wing X'foiia^p". V=ica? 

 mature axis of Kanlfussia ; but it is dorsi- ^ ' 

 ventral, and rather more elongated between 



the leaf-gaps, in accordance with its creeping habit. 4 In Angiopteris and 

 Marattia the final structure is more complex, though the initial steps are 

 similar. There is in their seedlings also a solid protostele : in the central 

 part of its xylem-core certain cell-rows cease to differentiate as tracheides, 

 but give rise to a parenchymatous pith : the siphonostele thus formed 

 becomes broken up by leaf-gaps, thus giving rise to a dictyostelic cylinder 5 

 (Fig. 291). Subsequently, as the stem passes to maturity, there arise 



1 But Jeffrey (Phil. Trans., 1892, I>, vol. cxcv., p. 120, etc.) states that in several 

 species of Danaea the stele is tubular in the seedling, and that it is interrupted by leaf-gaps. 

 That may be so in older conditions, and Jeffrey's material does not appear to have 

 been young enough to decide the question for the earliest stages. 



-lirebner, .-///;/. of Bot., xvi., p. 524. 



:! Gwynne-Vaughan, Ann. y Bot., xix., p. 259. 4 Kiihn, I-lora, 1889, p. 475. 



farmer and Hill, Ann. of Bot., xvi., p. 371. 



Kl',. 2QI. 



