XIX] COMPARISON 89 



HelmintJwstachys with their definitely circumscribed steles, their early 

 protostely and subsequent medullation, and their uniformly undivided, and 

 occasionally Clepsydroid leaf-trace, show clear anatomical relations with 

 other early Fern-types, such as the Botryopterideae and Fossil Osmundaceae. 

 More especially the slight secondary thickening of Botrychium finds its 

 analogue in the Palaeozoic Botrychioxylon, and occasionally also in the 

 Zygopterideae. On the other hand, the laxer stele of Ophioglosswn, with its 

 early loss of the endodermis and its expansion into an open dictyostele of 

 unsheathed strands, presents an advanced state that reaches a climax in the 

 distended sappy stem of O. palmatuni (Fig. 385). Further, the divided leaf- 

 trace of O. palmatum, pendulum, and simplex points out those species as 

 anatomically advanced (Figs. 355, 356). On these grounds it may be held that 

 Botrychium and Helminthostachys are relatively primitive members, while 

 Ophioglossum (and especially § Ophioderma and § Cheiroglossd) will be 

 regarded as the most advanced genus of the family. 



How then do the facts relating to the fertile spike accord with this 

 conclusion? How shall the branching and multiplication of spikes seen in 

 Ophioglossum be harmonised with the idea of its being a relatively late type 

 of the family? It has been the habit to regard Ophioglossum as primitive 

 chiefly because its spike appears to be the simplest in form. Is this conclusion 

 justified? It may be tested by comparison with known Palaeozoic fossils. In 

 theCoenopteridaceae we see Ferns having sporangia of a type not far removed 

 from those of Botrychium and Helminthostachys. The way in which those 

 sporangia are borne, marginally or distally, accords with experience among 

 the fossils. It requires no great strain of imagination to see in the unique 

 spike of Helminthostachys a condensed form of such a type of sporophyll as 

 that of Stauropteris: and this comparison is supported by the fact that a 

 vascular strand passes to the base of each sporangium. On the other hand, 

 the sporangium of Ophioglosswn itself is not of a type known among fossils, 

 any more than is the spike which bears it. 



The suggestion that Ophioglossum is relatively late and derivative implies 

 a progressive sinking of the sporangium into the tissue of the part that 

 bears it, accompanied by an enlargement of the sporogenous tissue, and of 

 the spore-output. No doubt it will be remarked that this is contrary to the 

 general evolutionary progression in Ferns, so clearly brought out by the 

 spore-enumerations, and by the comparisons of sporangial structure described 

 in \"ol. I, Chapter XIII. These demonstrate for Ferns at large the progressive 

 decrease in size of the sporangium, and fall in the spore-output. This is 



im Blattaitfbati, Jena, 1922. The views which he advances appear to accord with the general argu- 

 ments of this volume, and in particular with the opinion here expressed as to the venation in the 

 Ophioglossaceae ; but they will find more detailed application in the treatment of the more advanced 

 Filicales to be given in Vol. ni. 



