i6o SCHIZAEACEAE [CH. 



medullary region in several species both in young and in adult stems, the 

 facts appear to favour a theory of progressive amplification of the stele 

 with medullation in Schizaea, rather than one of reduction. (Dr J. M. 

 Thompson, Trans. Roy. Soc. Edin. lii, p. 715, 1920: here the question is 

 discussed, and the literature fully quoted. See also Vol. I, Chapter Vll.) 



In the genus Anemia the Section Anemiorrhiza has a typical solenostelic 

 structure in its creeping stem {A. adiantifolid) : but in upright species, where 

 the axis is short and the leaves are more crowded, the stem contains a well- 

 developed dictyostele, the foliar gaps overlapping. Not only is this so, but 

 their stems may be perforated by deep depressions of the surface at the axil 

 of each leaf, giving that curiously ventilated basket-like structure described 

 by Gwynne-Vaughan. This is particularly well shown in A. hirsuta, and it 

 also appears in A. Phyllitidis (see Vol. I, pp. 1 49-1 50). The stem of JMohria 

 is also dictyostelic, but without these involutions, while the pith and cortex 

 contain dense masses of brown sclerenchyma. This with the protection by 

 broad dermal scales are features related to its xerophytic habit. Mohria may 

 be held structurally as the most advanced type of the Family, which thus 

 illustrates all steps from protostely to fully elaborated dictyostely. In 

 stelar features Lygodiwn may be regarded as the most primitive, ScJiizaea 

 as holding a middle position, while Ajieniia and Mohria are the most 

 advanced. 



The stelar advance is carried out with a consistently undivided leaf-trace 

 of a very characteristic structure, as seen in Anemia and Mohria (Fig. 445). 

 It may be regarded as based upon the underlying C-shaped trace of the 

 Osmundaceae. The meristele has a peculiar saddle-shaped outline, much 

 attenuated in the middle region (compare Loxsoma, Vol. I, Fig. 158). There 

 are four exarch protoxylems, and the metaxylem is much enlarged right 

 and left: it may be extended at the margins into adaxial hooks. In 

 Lygodium the petiolar trace is on the same general plan, but it is contracted 

 as already noted. In Schisaea the trace is peculiarly three-lobed, but with 

 the protoxylems as before on the adaxial face (Fig. 165, b. Vol. l). The 

 interest of these leaf-traces lies first in their constantly undivided state: 

 next in the fact that those of Anemia and Mohria conform closely to those of 

 certain derivative Marginales {Loxsoma, Plagiogyria, Davallia speluncae, 

 Fig. 159, Vol. l): thirdly, in the peculiar modifications to which they are 

 liable in specialised leaves such as Lygodium and Schizaea, which are still 

 referable nevertheless to the same general scheme. 



Lastly, the venation deserves to be noted. In Schizaea it is simple, or 

 repeatedly and obviously dichotomous, with blind endings and no vein- 

 fusions. The whole blade presents a near approach to the primitive equal 

 dichotomy, being more or less completely webbed. Mohria also has the 

 venation always open, but here the strongly dichopodial branching makes 



