VIII] 



SOLENOSTELY 



145 



The validity of the conclusions based on these ontogenetic 

 facts relating to one of the Superficiales, and the general nature 

 of the facts themselves, may be tested by examination of relatively 

 primitive types of the Marginales, which are believed to represent 

 a parallel evolutionarysequence, though phyletically distinct from 

 the Superficiales. The Schizaeaceae illustrate the steps of stelar 

 elaboration better than any other family of Ferns. Lygodiuvi is a 

 permanently protostelic type : Schisaea itself has been described 

 above, as illustrating advances in medullation, and the formation 

 of xylic gaps and pockets (p. 133). In Anemia^ % Attetniorrhiza^ 

 solenostely is seen, while § Eu-A?iemtaa.nd § Mohria2Lr& examples 

 of dictyostely. But they all have relatively short axes. For the 

 demonstration of the steps leading to a typical solenostely a Fern 

 with long internodes is preferable. It is found in Loxsotna, a 

 genus of Schizaeoid-Dicksonioid affinity, already described 

 anatomically by Gwynne-Vaughan as an example of typical 

 amphiphloic solenostely, showing a foliar gap immediately 

 above the point of departure of each gutter-shaped leaf-trace 

 (Fig. 130). The ontogeny may here again be expected to throw 

 light upon the probable origin of solenostely. 



It starts with a basal region of protostely, which soon passes 

 into the condition where a medulla is present surrounded by a 

 ring of xylem (Fig. 135). Internal phloem makes its appearance 

 with rather irregular arrangement, and extends for a considerable 

 distance, giving the so-called Liitdsaya-cond.ix\on. As the leaf- 

 traces pass off in this region {b—f) pockets of variable depth are 

 formed, with involution of the endodermis. These extend into 

 the bulky stelar pith in somewhat the same way as in Schizaea 

 dichotoma. Passing upwards a point is reached where a pocket, 

 widening upwards, opens not only into a single foliar gap, but 

 extends onwards into a succession of them {g). Here, as in Gleich- 

 enia pectifiata., a continuous internal endodermis is established, 

 surrounding a central column of pith, which becomes sclerotic 

 upwards, and is connected directly through the successive foliar 

 gaps with the cortex. There is again no indication of sliding 

 growth, or intrusion of tissue from without. All the vascular 

 tracts are completely shut in by an unbroken endodermal sheath. 

 Here also the structural facts as actually presented in a series 

 of transverse sections point to changes of procambial destina- 

 tion of the cells /« situ, so as to form parenchyma, sieve-tubes, 

 or endodermal cells in place of tracheides present in earlier stages 

 of the individual. A statical change of quality of the cells pro- 

 duced will account for the transition actually observed, from 

 protostely to solenostely, better than any suggestion of dynamical 

 intrusion from without. The steps of the final transition to sole- 

 nostely in Loxsoma are shown in Dr Thompson's reconstruction 

 (Fig. 135), and they resemble those seen in Gleichenia pedinata 

 so closely that together they may be accepted as illustrating 

 for Ferns generally the structural passage from the protostelic 

 to the solenostelic state. 



F'g- '35- Plan of the stelar 

 construction in the upper 

 part of a juvenile plant of 

 Loxsoma Cioiiniig/iamu, 

 as seen in median section. 

 It starts from the medul- 

 lated stele at the base, and 

 attains a full solenostelic 

 structure in its upper part. 

 a—f= leaf- bases ;^ — g'^ = 

 leaf-gaps; ;5= "pocket." 

 Fordetailsseetext. (After 

 Dr M'=Lean Thompson.) 



