KUNGL. SV. VET. AKADEMIENS HANDLINGAR. BAND 57. N:0 I. 37 



mechanical tissue, though the preservation is not such as to enable us to draw any de- 

 finite negative conclusion. The fact that the central strand has been able to cause an 

 impression in the matrix through the thick surrounding tissue, however, proves that the 

 latter was of soft texture and probably parenchymatous. It is clear, at any råte, that 

 the vascular strand plays a very insignificant part in the construction of the stem as 

 compared with the softer tissues. 



2. The probable absence of secondary growth in Arthro- 

 stigma. The stems never exceed a certain constant thickness, and even in the thickest 

 stems the vascular strand remains equally narrow in relation to the surrounding soft tissue. 



3. The irregular disposition of the leaves, as related in the 

 descriptive part. 



4. The small size of the leaves, as compared with the 

 thickness of the stems, and t h e i r resemblance to emergen- 

 c e s in Psilophyton princeps. The leaves of Arthrostigma are sometimes a little longer 

 than the diameter of the stem, sometimes considerably shorter. The spines of Psilophyton 

 princeps are usually shorter than the diameter of the stem. It is difficult to imagine that 

 even the leaves of Arthrostigma were sufficient to provide alone for the assimilation. The 

 lateral appendages in Psilophyton princeps are so small as compared with the stems that 

 it seemed absurd to call them leaves, and they ha ve therefore been termed spines in the 

 descriptive part. If the close analogy with Arthrostigma seems to render it necessary 

 to regard them as rudimentary leaves, their habit of mere emergences is all the more 

 remarkable. The fact that among the stem-like structures abounding in these oldest 

 land-floras it is often difficult to decide whether the lateral outgrowths should be re- 

 garded as leaves or emergences would in itself appear to be significant. 



Evidence of the existence of megaphyllous Pteridophyta. 



It seems to be a general opinion that large fern-fronds with typical pinnules reach 

 at least as far back in the geological history as the microphyllous forms. This opinion 

 is based on a review of the Upper Devonian floras, in which the genus Archaeopteris 

 occurs side by side with for instance forms of Bothrodendron. If the discussion is extended 

 to the Lower Devonian floras, however, as typically developed at Gaspé and at Röragen, 

 the question becomes more complicated. 



At first it is necessary to state that neither in the Lower Devonian of Gaspé nor at 

 Röragen are there found any forms of Archaeopteris or similar types. Dawson (1882 a, 

 p. 113) remarks on the absence of ferns in the Lower Devonian of Gaspé, and in his dis- 

 cussion of the contemporaneous beds of Scaumenec and Campbellton lie says: »It is 

 evident that here, as in Gaspé, Prototaxites and Psilophyton with Arthrostigma are cha- 

 racteristic of the Lower Erian, while ferns of the genus Archaeopteris are equally so of 

 the Upper Erian immediately underlying the Carboniferous ». The most obvious negative 

 characteristic of the Röragen flora and, according to Dawson, also of the Gaspé flora, 

 is indeed the absence of any fern-pinnules or other f oliar laminae. On the other hand 



