38 T. G. HALLE, LOWER DEVONIAN PLANTS FROM RÖRAGEN IN NORWAY. 



there occur abundantly remains which might well be held to be rachises of fern-fronds 

 or their branches, were it not for the absence of any traces of laminae. This feature 

 might be regarded as a result of the imperfect preservation, but its constancy in different 

 localities renders this explanation insufficient. As a matter of fact, the preservation of 

 the Röragen plants is rather good, especially in the slate: this is shown sufficiently by 

 the delicate structures figured in the plates accompanying this paper. If the absence 

 of laminae of fern-fronds is explained by the poor state of preservation one cannot but 

 complete the circle with the admission that the chief reason to regard the preservation 

 as poor would be just this absence of foliar laminae! On the whole, the present evidence 

 seems to indicate that the Lower Devonian flora is characterized by the absence of 

 flattened fern-pinnules or foliar laminae generally. 



On the other hand it cannot be denied that many of the branching structures of 

 the types designated as »Hostimella » or ^Aphyllopteris » strikingly recall rachises of fern- 

 fronds. Especially among the large, freely branching fronds of the Lower Carboniferous, 

 as for instance those figured by Stur (1875 — 77, pls. 26, 27, 33 — 35), close analogies may 

 be founcl. Sometimes such flattened branch-systems themselves ha ve an aspect similar 

 to that of finely divided fronds: so, above all, the specimen in pl. 2, fig. 7, discussed above 

 under the generic name of Hostimella. In the form deseribed as Psilophyton Goldschmidtii 

 such branches arise from a common rachis or axis, which however bears spines similar 

 to those on the stems of Psilophyton princeps. Psilophyion Goldschmidtii habitually 

 rather suggests a large frond, but it is difficult to draw any distinction between its axis 

 or rachis and the stems of Psilophyion princeps, which in their turn show a close relation 

 to Arthrostigma and undoubted lycopodiaceous stems. 



It might perhaps be suggested, although this is pure speculation, that megaphyl- 

 lous forms may be evolved from a type like Psilophyion Goldschmidtii. The lateral 

 branches of this form already appear to have a bilateral or dorsiventral symmetry. 

 The rapid tapering of the segments of isolated branch-systems similar to the lateral 

 branches of Psilophyion Goldschmidtii suggests a limited growth in some cases. Such lateral 

 branches of limited growth may be imagined to develop laminae by a process of »cladodi- 

 fication». A similar development has been suggested by the late Professor Lignier 

 (1903, 1908 — 1911) in his interesting speculations on the first evolution of the different 

 branches of the pteridophytic stock. Lignier even used Psilophyton princeps as a 

 starting point. He adopts the view of a diphjdetic origin of the leaves of the Pteridophyta, 

 starting from a »prohepatic type», derived from the algae. The leaves of the Lycopo- 

 diales are distinguished as »phylloids» and regarded as developed phylogenetically by 

 »enation» in the manner of emergences. The fronds of the megaphyllous forms, on the 

 other hand, are true leaves formed by different iation of thallus-branches in accordance 

 with the current opinion. It would seem that what little is known at present about the 

 Lower Devonian flora is well in accord with Lignier's views. We have in Psilophyton 

 princeps, imperfectly though it be known, a plant which has actually existed and which 

 answers well to the type theoretically required as a starting point. Similar plants, with 

 well developed stems and small lateral appendages which may be compared to either 

 emergences or leaves, were dominant in the Lower Devonian flora; and there is reason 



