124 BRITISH EOCENE FLORA. 



kernel of which contains a bulky embryo, furnished with two unequal cotyledons. The 

 stem has a relatively large pith, and by this, as well as by its punctated woody fibre and 

 septiferous bark and some other characters, its relation to Cycads is apparent. The 

 Ginkgos can be traced back to the Carboniferous through a series of increasingly modified 

 forms, even those from the Permian being not dissimilar in aspect to those which still 

 exist. Prof. Williamson is convinced that Dadoxylon was a closer ally of Ginkgo than of the 

 Cycads, the leaf-bundles being in pairs, a peculiarity of construction now wholly confined 

 to the former. The type was by no means isolated in the Carboniferous as at present, 

 for many varieties of Dadoxylon, and fruits resembling those of Ginkgo are known, and 

 its ancestral form Ginlcgophyllum approaches both Noegyerathia and Corddites, as well as 

 the most ancient Cycadaceae. The course of evolution does not demand that all Coniferse 

 should have descended from Ginkgo as a common ancestor, but the living species is 

 certainly the only one which, apart from its exogenous trunk, can convey to us a faithful, 

 if weakened, image of the Coniferae as they existed in times anterior to the organic 

 changes which transformed some into cone-bearers and others into Taxeae. 



It only remains now to trace as far as possible the evolution of the Conifers from 

 the primitive type. The oldest examples known to us possessed an exogenous 

 growth. Their leaves were broad and attenuated at the base with parallel dichotomising 

 venation. The male and female inflorescence were separated, and formed spikes composed 

 of bracts or metamorphosed leaves arranged round a central axis. The subsequent modi- 

 fications have been in the direction of the successive reduction of the breadth of the leaf 

 and number of veins, and their combination into a mid-rib ; and it is amongst the most 

 primordial genera, Araucaria and Agathis, that the largest leaves are to be found. 



The Taxe^e are older as a group than the true Coniferae. During the progressive 

 development of the Coniferae to their present state, great numbers of types must have been 

 evolved which have since entirely disappeared. A few, more persistent than others, have 

 come down to us and seem to reveal at least some of the stages by which the cone- 

 bearing Coniferae were differentiated from the rest. There has been a tendency in the 

 Taxeae for the floral axis to diminish, and in the other orders for it to increase. One 

 well-nigh constant element in the Taxeae and the Podocarpeae is the swelling of the sub- 

 stance surrounding the base of the ovule until it forms either a membranous or fleshy 

 cup. Dacrydium, Podocarpus, P/tyllocladus, and Taxus are diverse examples of this. 

 Saxegothea is perhaps the highest form of Taxad, for the bracts form a sort of cone, each 

 scale of which bears an inverse ovule at its base, seated in a membranous cup. In the 

 Yew the male flower is made up of a number of bracts, very like the inflorescence of 

 Equisetum, as pointed out by Sachs ; while the female flowers "spring from the axils of 

 foliage-leaves belonging to elongated woody shoots and have the form of short branches 

 covered with decussate scale-like bracts," the axis of the shoots ending in a terminal 

 ovule. 1 The last is not dissimilar, according to Saporta and Marion, to that of the Abie- 



1 Sachs, p. 515. 



