232 A GENERAL REVIEW OF THE SIMPLICES [CH. 



The distinction thus to be applied systematically depends for its value 

 upon the high degree of constanc)' of the character used, and upon the 

 relatively early establishment of it as shown by the fossil evidence. The 

 leading types of the Superficiales have possessed superficial sori since 

 Palaeozoic times. But the distinction is not absolute. As has been fully 

 explained in Vol. I, Chap. XII, the passage of the sorus from the margin to 

 the surface of the broadening sporophyll may be followed in the individual 

 development among the Schizaeaceae of to-day, while comparison of living 

 species and genera in the Dicksonioid-Davallioid series demonstrates a 

 similar change of position as not unfrequent in them. The first impulse may 

 naturally be then to rule out this distinction as systematically worthless. 

 But the point is that in certain Families the event happened as early as Palae- 

 ozoic time, and the result of it has remained constant ever since ; in others the 

 originally marginal position Jias been retained with high constancy, though in 

 some instances it is even now in course of being departed from. The former 

 state is what we see in the Superficiales as exemplified by the Gleicheniaceae, 

 and the same holds also for the Marattiaceae : the latter is what we see 

 in the Marginales as exemplified by the Schizaeaceae, and the Ophioglos- 

 saceae retain persistently that primitive character. Ultimately both sequences 

 are probably based upon a prevalence of narrow leaf-segments in the remote 

 ancestry, with a distal sporangium borne upon the tip of each, as seen in 

 Stauropteris. Webbing would result in a flattened leaf-surface wath marginal 

 sporangia: further widening, and slipping of the sporangia, earlier or later 

 in descent, to the lower surface would give the state so frequently seen in 

 modern Ferns. In the classification here adopted these Ferns will be ranked 

 systematioally according as they retained the primitive marginal position, 

 or departed from it early or late in the history of descent. 



Professor von Goebel holds that this distinction is an artificial one, in 

 particular as applied to the Dicksoniaceaeand Cyatheaceae iOrganographie, 

 ii, p. 1 194, 1918). I hope that that opinion may be reconsidered by my 

 esteemed colleague, when he takes into full account the palaeontological 

 and anatomical facts here detailed, together with the high degree of con- 

 stancy of the one soral position and of the other in the large series related 

 on the one hand to the Schizaeaceae, and on the other to the Gleicheniaceae; 

 also the very gradual steps of emergence of later terms of these series in 

 relation to those very ancient stocks. It will then appear that the distinction 

 is supported by a very great body of fact, and by wide comparisons. The 

 stigma of artificiality cannot justly be placed upon the results of such wide 

 comparative argument. The distinction according to soral position would 

 rather appear to be a Natural Classification in the fullest sense of the word. 

 It will be maintained and elaborated in this work, due attention being given 

 to those exceptions upon which adverse criticism may be founded. Relatively 



