xxxviij COMPARISON 39 



it the formation of coenosori, and the soral fusion is seen as it were tentatively 

 in several distinct genera: this maybe held as a homoplastic result produced 

 in a plurality of phyletic lines. The fact that it is seen in certain species of 

 Nephrolepis {N. dicksonioides and acutifolia, Fig. 592), illustrates the point in 

 a genus in which the sori for the most part retain their individualit}-. 



Difficulties are inherent in any classification which aims at being natural, 

 that is evolutionary. In the present instance it may be held that the three 

 sections of the Davallioid derivative Ferns exhibit states or conditions arrived 

 at probably from a common origin, but along a plurality of individual lines. 

 They would thus rank theoretically with such states or conditions as were 

 expressed by the definitions of the old genera Acrostic/mm or Polypodinm: 

 these can no longer be maintained in their old classificatory sense, since both 

 states have undoubtedly been attained polyphyletically. The difificulty thus 

 encountered in handling this first large group of advanced Leptosporangiate 

 Ferns is inherent in greater or less degree in all other large groups. It need 

 not oppress the mind unduly so long as the position is clearly understood. 

 A complete artificial classification is always possible, and is indeed necessary 

 for floristic use. A complete phyletic classification will only become possible 

 with complete knowledge of the descent of the organisms classified. The 

 second cannot replace the first under present conditions, owing to the 

 imperfection of present knowledge. But it can lead to a correction and 

 amendment of classification for floristic use, so as to make it run ever more 

 nearly along the lines of probable evolution. This is what appears to result 

 from the suggested grouping of the Davallioid Ferns, as compared with the 

 catalogues of the genera of this group given elsewhere. 



BIBLIOGRAPHY FOR CHAPTER XXXVII 



593. Hooker. Genera Filicum, Plate LXii, Dictyoxiphiuin. 1842. 



594. Hooker. Species Filicum, Vol. i, p. 150. 1846. 



595. KuHN. Die Gruppe der Chaetopterides. Berlin. 1882. 



596. De Bary. Comparative Anatomy, Oxford, pp. 287, 347. 1884. 



597. Lachman. Nephrolepis^ Ann. Soc. Bot. de Lyon. 1888. 



598. Prantl. Das System der Fame. Breslau. 1892. 



599. Christ. Farnkrauter, p. 289. 1897. 



600. Bower. Studies in Spore-Producing Members, iv, Phil. Trans, p. 75. 1899. 



601. DiELS. Natiirl. Pflanzenfam. I, 4, p. 204. Also p. 139, where the systematic literature 

 is fully cited up to 1902. 



602. Gwynne-Vaughan. Solenostelic Ferns, 11, Ann. of Bot. xvii, p. 689. 1903. 



603. Van Roseburgh. Malayan Ferns, Batavia, pp. 255, 567. 1909. 



604. Bower. Studies III, Ann. of Bot. xxvii, p. 443. 1913. 



605. Gwynne-Vaughan. Climbing Davallias, Ann. of Bot. p. 495. 1916. 



606. Bower. Studies VII, Ann. of Bot. xxxii, p. i. 1918. 



607. Von Goebel. Organographie, 2te Aufl. p. 1143. 1918. 



608. Von Goebel. Prosaptia, Ann. du Jard. Bot. de Buit. xxxvi, p. 148. 1926. 



