126 



DRYOPTEROID FERNS 



[CH. 



phylum which, according to the argument advanced here, is a coherent 

 derivative of the Superficiales, springing from the Cyatheaceae, viz. the 

 Aspidieae. It is well represented in the genera Diacalpe and Peranema, 

 which link them with the Cyatheaceae. Though the lines of occurrence of 

 the perispore are as a whole somewhat blurred by enigmatical exceptions, 

 the facts give substantial support to the view here expressed, viz. that the 

 Aspidieae having a perispore are a coherent phylum with no direct relation 

 by descent to the Marginales, in which a perispore is absent. 



Fig. 658. Spores of Aspidiuni trifoliatum, after Hannig. 

 (7, /' = ripe spores with prickly exospore {ex) and trans- 

 parent perispore [psp), appearing like a loose sac (x 500). 

 c=:part of the exospore and perispore more highly 

 magnified. 



The gametophyte of the Male Shield Fern presents no features of impor- 

 tance excepting the presence of short unicellular hairs on the margin and 

 lower surface (Fig. 19, Vol. l), such as are present on the prothalli of Woodsia 

 and Diacalpe: these are held by Von Goebel as transitional to those of the 

 Cyatheaceae {Organographies II, 2, p. 959, Fig. 952). 



The lid-cell of the antheridium is undivided, according to the ordinary 

 Leptosporangiate type. 



The facts and comparisons thus brought forward confirm the position of 

 the Male Shield Fern as a derivative from a Cyatheoid ancestry, with the 

 VVoodsioid genera Diacalpe and Peranema as synthetic intermediate types. 

 These together appear as later steps in the sequence of Ferns which bear 

 consistently superficial sori. That sequence leads upwards from the Gleicheni- 

 aceae, with their simple radiate, non-indusiate sori of a type prevalent in the 

 Mesozoic period, and represented even in Carboniferous times. The steps have 



