620 



FILICALES 



types of leaf, to which that of Dipteris is related by D. Lobbiana', but 

 from this simple narrow-leaved type, with its single row of sori on either 

 side of the midrib, the broader-leaved Dipteris has broken away as its leaf- 

 area enlarged, and the sori have been spread over the extended surface, 

 while the absence of their individuality gives the key to the way in which 



FIG. 344. 



Dipteris Lobbiana (Hook.), Moore. 

 Parts of two pinnae, showing narrow form, 

 venation, and regular disposition of the 

 sori. Natural size. 



FIG. 345. 



From a specimen collected by Capt. Hope, R.N., 

 on the " China Station," but without exact locality: 

 recognised as Dipteris qidnquefurcata, Baker. Pinna 

 showing greater width, and sori arranged below in two 

 lateral series as in D. Lobbiana, but spreading out 

 upwards, with many fissions, over the more extended 

 surface. Natural size. 



the result has been brought about, viz. by fission. This process, so clearly 

 seen in the few species of this very natural genus, has probably occurred 

 also in other types of Ferns. It is suggested by Kaulfussia among the 

 Marattiaceae, but much more obviously in various lines of the Polypodiaceae. 

 It will have to be reckoned with in any general conception of the phylogeny 

 of the leaf in Ferns. 



Examining the sorus itself, it is composed of a number of sporangia, 

 and numerous glandular hairs are associated with them. The sporangia 



