FERNS— SUMMARY 351 



The fossil history of the Osmundaceae from an 

 anatomical point of view is now being investigated by 

 Kidston and Gwynne-Vaughan. Among the forms 

 already dealt with, the Osmundites skidegatensis of 

 Penhallow, from the Lower Cretaceous of Queen 

 Charlotte Islands, British Columbia, may be mentioned. 

 The preservation is astonishingly perfect and the stem 

 is remarkable for possessing a well-developed internal 

 phloem. This feature, however, was found to be 

 absent from the still older members of this family. 



There appears to be little satisfactory evidence as 

 yet as to Mesozoic Hymenophyllaceae. 



The family Dipteridineae, formerly included under 

 Polypodiaceae, appear to have a long geological 

 record, Ferns referred to this group having been found 

 in the Lias, Rhaetic, and Trias. 1 



True Polypodiaceae have not been traced further 

 back at present than to Jurassic rocks. 2 



The palseontological data just cited may be 

 compared with the arrangement of the families of 

 Ferns, proposed by Professor F. O. Bower, 3 on the 

 basis of an exhaustive investigation of the development 

 of the sorus, and the output of spores. The main 

 lines of his classification are exhibited in the following 

 table : — 



1 Seward and Dale, "On the Structure and Affinities of Dipteris, with 

 Remarks on the Geological History of the Dipteridineae," Phil. Trans. 

 Royal Soc. B, vol. 194, 1901. 



2 For a fuller summary see Zeiller, " Les Progres de la Paleobotanique 

 de l'Ere des Gymnospermes," Progressus Rei Botanicae, vol. ii. 1907. 



3 " Studies in the Morphology of Spore-producing Members," iv. The 

 Leptosporangiate Ferns, Phil. Trans, vol. 192, B, 1899. For a more recent 

 statement, see his Origin of a Land-Flora, 1908, especially the scheme 

 on p. 653. 



