﻿2 SCOTT, Early History of Seed-bearing Plants. 



unknown (at least until within the last year or two) ; 

 in a few cases such evidence as was available indicated a 

 type of fructification very different from anything found 

 among true ferns. 



In all these numerous instances there were thus no 

 real grounds for referring the plants in question to the 

 Filicineae, beyond the characters afforded by the frond, 

 characters which, by themselves, are notoriously untrust- 

 worthy. 



Although Stur, in 1883, excluded certain of these 

 genera (e.g. y Aletkopteris, Neuropteris, and Odontopteris) 

 from the ferns, for the reasons indicated, the majority of 

 palaeobotanists were content to leave them in their 

 traditional position, preferring to rely on positive char- 

 acters however inconclusive, rather than on evidence 

 which as yet was almost wholly negative. And in this 

 they were not without justification, for the absence of any 

 fern-fructification could always be provisionally accounted 

 for by the hypothesis of dimorphism. Dimorphic ferns, 

 or at least fern-like plants, are common in the Carboni- 

 ferous Flora, and in cases where no Filicinean fructifica- 

 tion was found on the ordinary leaves it was permissible 

 to suppose that it might have been borne, as in Osmunda, 

 for example, among recent ferns, on a specialized fertile 

 frond, or portion of a frond, differing too widely from 

 the sterile form to admit of identification. 



Positive evidence as to the nature of some of the 

 doubtful ferns of the carboniferous first came from 

 anatomical investigation. When, just fifteen years ago, I 

 paid my first visit to the late Prof. Williamson, in his 

 house at Fallowfield, and there became acquainted with 

 the treasures of his collection, the plant which perhaps 

 most impressed me by the beauty and interest of its 

 marvellously preserved structure was Lyginodendron old- 



