246 Scientific Proceedings, Royal Dublin Society. 



particularly as Dr. D. H. Scott had expressed the opinion to me verbally that 

 Archaeopteris would turn out to be a Pteridosperm and not a Fern. The 

 quarry has yielded many examples (PI. IX, fig. 1) of the stipulate leaf- bases 

 first described by Kidston (6). The base of attachment is broad, grooved 

 (PI. IX, fig. 2), and makes a constant characteristic angle (PI. IX, fig. 3) with 

 the rachis proper, which is swollen on its adaxial side proximally (PI. IX, 

 fig. 4, and fig. 2, PL XIV). Comparison may be made with the swollen base 

 of the petiole of the Pteridosperm — Rhetinangium Arberi— which Gordon (7) 

 shows is due to excessive development of cortical parenchyma, and forms 

 a weak tearing region. Several slabs show short thick stumpy stems bearing 

 spirally arranged stipulate leaf-bases like those of Archaeopteris. The figure (5), 

 PL IX, shows at A such a stem, and lying on the slab another slab which at 

 B shows a known leaf-base of Archaeopteris which is externally suggestive in 

 its basal part of that of Angiopteris or other modern Marattiaccous plant. 

 It does not appear, however, to have ended apically in a rosette of fronds. 

 Figure 1, PL X, which represents a stem, apparently belonging to the same 

 fossil as Archaeopteris, suggests that the Archaeopteris stem was elongated 

 apically and bore distant fronds — that possibly it was of a straggling if not 

 climbing habit. This Kiltorcan stem of Archaeopteris reminds one very 

 forcibly of the illustration Dawson (8) gives of his tree-fern Caulopteris 

 Lockwoodi from the Upper Devonian beds of Gilboa, New York. It is not 

 improbable that this Caulopteris is the stem of Archaeopteris, the foliage of 

 which is recorded from the same deposits. 



True Archaeopteris had, so far as we know, an unforked stipulate tripinnate 

 frond with sessile pinnules borne directly on its rachis. The veins are dicho- 

 tomous and numerous. There is no midrib in the pinnule as in Sphenopteris 

 usually ; two or three bundles enter the rachidule from each pinnule, as is 

 also the case for each pinna at its point of insertion on the rachis. Apparently 

 four, five, or more bundles enter the stem from the rachis, whether separately 

 or, as in Rhetinangium, as one continuous band I cannot say. 



On the distal side of the leaf-base cushion on the adaxial surface there 

 is to be seen in some of the impressions a forked appendage represented in 

 fig. 1, PL X. This appendage may be connected with reproduction. It 

 suggests comparison with the abaxial fertile appendages discovered by 

 Nathorst (9) in Ccphalotheca and with the basal appendage in Heterangiitm 

 hibernicum (10). Its occurrence made me hesitate to refer such specimens to 

 Archaeopteris, where forking is a rarity as compared with Sphenopteris Hookeri. 



Though the sterile pinnules are not lobed, the fertile ones are sometimes 

 markedly so to such an extent as to deserve the term " digitiform." This 

 occurrence of segmentation in the fertile pinnule ought not to be overlooked 



