E. A. N. Arber—The Tria-^.sic Plant Yuccites. 13 



the axis. In the light of Mr. "Wills' new discovery, I am now of 

 opinion that the mode of attachment here, though not very clearly 

 seen, is probably spiral, and that we are not dealing in this case 

 with a pinnate frond. 



Other fossils of u similar type are to be found in the Cretaceous 

 Krannera mirahilis of Velenovsky and the Eolirion {^Elolirioii) primi- 

 genium of Schenk.^ The latter strongly recalls the Bromsgrove 

 specimens, and in Schenk's example the leaves are obviously attached 

 spirally to an axis or shoot, with which many of them appear to be in 

 continuity. AVe have, therefore, a family of plants including the 

 genera Yuccites, Schimp. & Moug., Palceozama, Phillips, Krannera, 

 Vel., and Eolirion, Schenk, existing in both Triassic, Jurassic, and 

 Wealden times, with large Monocotyledonous-like leaves, spirally 

 arranged on the shoot. The question is, are these fossils best regarded 

 as Monocotyledons, or as members of the Cordaitales or Coniferales, 

 or again as a distinct family possibly intermediate between the 

 Cordaitales and Coniferales ? For a complete answer we must wait 

 until the fructification is known. At present the balance of 

 probability does not appear to me to lie more in one direction 

 than in another, if we exclude the possible attribution to the 

 Monocotyledons, which, on general grounds, is by far the least 

 probable. • It is generally agreed that Monocotyledons did not exist 

 before Lower Cretaceous times, and that the earliest remains of 

 this group come from rocks of slightly more recent age than the 

 British Wealden. 



With regard to the possible Cordaitean affinity, there is much to be 

 said in its favour, as Zeiller - has already pointed out. The fact that 

 no ArtisiaYxk^ casts have been collected from the Keuper of 

 Bromsgrove is hardh' trustworthy evidence, either for or against the 

 argument. They are already known from a higher horizon, and 

 negative evidence is rarely reliable if unsupported by other data. If 

 the attribution to the Cordaitales proves to be correct, then this 

 group must have been most remarkable in regard to its long persistence 

 in geological time. It occurs in the Lower Carboniferous, if not in 

 the Devonian, and in Upper Carboniferous and Permian times it 

 formed one of the dominant elements of the flora. The argument that, 

 because most of the great dominant Palaeozoic groups had either died 

 out, or sunk to the position of subsidiary or insignificant factors in the 

 flora by the time we reach the Rhsetic, the Cordaitales were unlikely 

 to have persisted throughout Mesozoic times, is not, however, in itself 

 conclusive, if unsupported. The Leptosporangiate Ferns, for instance, 

 have maintained their place as dominant elements in the flora 

 throughout the Mesozoic and Tertiary periods even to the present day. 



Yet in the absence of any knowledge of the fructification I hesitate 

 to accept this view even provisionally. It is possible that the 

 fructifications of these Mesozoic plants may prove to be more 

 nearly related to the Coniferales than to the Cordaitales, or even 

 occupy an intermediate position between the two. The theory that 



' Schenk, Palccontogr., 1869, vol. six, Lief, i, p. 20, pi. vii fig. 4. 

 " Zeiller, ibid., pp. 212-13. 



