EOCENE FERNS. 69 



a frond and numerous smaller fragments. The section Euonodea, of the small genus 

 Onoclea, L., with which we have identified it, is represented by a single species, 0. sen- 

 sibilis, a large herbaceous dimorphic Eern, with copiously anastomosing veins, belonging 

 to the tribe Dicksonieae of the sub-order Polypodiacese. 



The pinnae in fig. 5 are simple and not divided to the rachis, the lobes bluntly 

 rounded and but slightly cleft, the margins undulated as in the existing fern ; their full 

 length is not preserved, the breadth is three quarters of an inch. 



To judge from existing species, the frond may have been of large size, with the pinnae 

 simple, decurrent near the apex, and slightly stalked towards the base. The membrana- 

 ceous fronds seem to have been quickly destroyed by maceration, for, with one exception 

 (fig. 5), only fragments (fig. 6) are known. They occur in a black shaly matrix, and 

 are therefore not easy to trace out, while the undulations of the pinnae have become pressed 

 into folds in places, rendering the venation confused. 



The rachis is moderately thin, the midribs of the pinnae slender, and the veinlets fine 

 and decided. The venation of fig. 5, accurately enlarged at 5 a, is the less common 

 type, and anastomoses less than is usual in the existing species. The primary veins 

 are seen to leave the midrib at an angle of about 60°. The veinlets are unequal and are 

 more numerous and important above the primary vein, or nearer the apex of the pinna 

 than below it. The lower veinlets start from near the base of the primary vein of each 

 group and unite with the next, forming a series of angulated areolations on each side of 

 the midrib. Nearly all fork once, sometimes twice, and are continuous to the margin 

 which is slightly denticulate. Some veinlets unite with those of the same, and others with 

 those of contiguous groups for short distances, and then again diverge, forming oblong, 

 angulate, or ovate meshes, which vary greatly in size. In fig. 6 a the anastomosis is more 

 frequent and complicated, the veinlets on the lower side of the primary vein being almost 

 entirely absorbed in the system of meshes proceeding from the upper side. This, by 

 far the more abundant type of venation, is reproduced exactly in the existing species. 

 Similar variations were noticed among the American fossils. 



Known from Mull, from the Fort Union group of Dakotah, and from Greenland,^ this 

 Fern differs strikingly from others of the European Tertiaries. It is found everywhere 

 in a lacerated condition, and presents great variability in its venation. Specimens 

 brought by Mr. Whymper from Greenland were also figured by Heer,^ and should now 

 be in the British Museum, though I am unable to find them. Their identity with the 

 Mull Fern is considered certain by Saporta -^ and if, as the plates show them to be, the 



^ 'Flora fossilis arctica,' 1868, vol. i, p. 86, described as Woodwardites arcticus from Atanekerdluk. 



2 ' Phil. Trans.,' 1869, pi. xl, fig. 6. 



^ The Marquis de Saporta, in a letter received since the text was in type, points out its resemblance 

 to Pleocnemia, a group of Aspidieae. 



" Votre Foug^re de Mull Filicites hebraidicus est effectivement ideutique avec le Woodwardites 

 arcticus de Heer ; il suffit pour s'en convaincre de comparer les figures du ' Flora fossilis arctica ' avec les 



