XXIIl] PSARONIUS 425 



are not represented in any Marattiaceous fern ; on the other 

 hand, they are comparable with the accessory strands met with 

 in stems of recent Cyathoaceous tree-ferns^ (cf. fig. 240). The 

 complex system of concentric dictyosteles is a feature more closely 

 matched in Angiopteris (Marattiaceae) than in any Cyathe- 

 aceous genus, the chief difference being in the more band-like 

 form of the steles in Psaronius, though in a stem of Angiopteris 

 figured by Mettenius we see a close approach to the extinct 

 type. The position of the protoxylem has unfortunately not 

 been clearly defined in Psaronius stems, but in P. Renaulti it 

 is stated by Scott'' to be endarch, a position which some of the 

 protoxylem strands occupy in Angiopteris^. The occurrence of 

 large sieve-tubes described by Scott in P. Renaultii is another 

 feature shared by recent Marattiaceae. In many of the conti- 

 nental species of Psaronius the phloem has not been preserved, 

 and our knowledge of this tissue is comparatively meagre. In 

 the Marattia;ceae the roots arise mainly from the inner portions 

 of the stele, while in Psaronius they are usually formed from 

 the external vascular bands. The formation of secondary cortical 

 tissue is a peculiarity of Psaronius; on the other hand, if 

 Butter worth* is correct in referring to that genus the roots 

 with secondary xylem, which he describes as P. Cromptonensis, 

 a comparison may be made with the occurrence of secondary 

 tracheae in the stem steles of Angiopteris^. 



The absence of mechanical tissue in the stem of Angiopteris 

 is in contrast with its occurrence in the fossil stems and in 

 recent tree-ferns ; but this is a character of secondary impor- 

 tance and one which can be readily explained by the difference 

 in habit between Angiopteris and Psaronius. 



The roots of Psaronius, more especially as regards the stelar 

 structure, are in close agreement with those of Marattiaceae. 



The reference to Marattiaceae of the great majority of fertile 

 fern-like fronds from Permian and Carboniferous rocks con- 

 stituted a strong a priori argument in favour of including 

 Psaronius stems in the same family, especially when it was 

 known that leaves with Marattiaceous synangia were borne by 



1 Eudolph (05). 2 Scott, D. H. (08). ' Shove (00). 



* Butterworth (00). " Farmer and Hill (02). 



