162 J. Starhie Gardner — Development of Dicotyledons. 



plants found in it. They require, before this can be done, and 

 deserve, the most profound study. 



"We have next the supposed Pi-e-Eocene or Pal-Eocene floras, the 

 chief of which are from the calcareous tuffs of Sezanne and the 

 limestone of Gelinden. Large collections from both these localities 

 have been in my possession, and are now in the British Museum. 

 The stratigraphical evidence as to the age of the former is inconclu- 

 sive, and the flora, rich in flowers as well as leaves, is unlike any 

 other. The Gelinden flora is mainly composed of small and simple 

 leaves, and is equally unlike any other, though its age seems to rest 

 on a surer basis. It does not appear so far that the age of any Pre- 

 Eocene or very old Eocene deposit could yet be recognized with 

 certainty through its Dicotyledons. 



We next come to the true Eocene deposits, and in the first place 

 I am able to correct the erroneous assumption that no flora in our 

 latitude comprising such modern and temperate-looking genera as 

 Platanus, Salix and Alnus can be Eocene. At Reading, underneath 

 the true London Clay with its typical marine fauna and the Reading 

 series, leaves and fruits abound of Platanus as completely differentiated 

 as existing species. With these are leaves and fruits of at least two 

 leguminous trees, a large-leaved tree with melon-shaped fruits of the 

 size of a walnut, and others forming a completely distinct and homo- 

 geneous flora, easily identifiable, and of perfectly defined age. It 

 strongly resembles the Dakota flora of America, but does not seem 

 known elsewhere in Europe. At Newhaven and at Coldam Com- 

 mon near Winchester, we have some Reading plants mixed with 

 others belonging to the somewhat higher and also perfectly charac- 

 terized horizon of Bromley, Lewisham, Croydon and Dulwich. 

 Platanus here mingles with many new forms of leaf, for the most 

 part strongly serrate, forming another group of plants quite unknown 

 out of England. We now reach the well-known flora of Sheppey, 

 led up to by the Heme Bay flora with Alnus Eichardsoni. The 

 Sheppey flora is extremely instructive, although it is so utterly 

 distinct from any other that it is of little value for purposes of com- 

 parison. Such vast numbers of fruits have been collected from it 

 that we may safely consider ourselves in possession of all the chief 

 types, yet, singular to say, though the presence of trees with such 

 easily recognizable fruits as Quercus, Fagus, Corylus, Castanea, 

 Juglans, and some Proteaceous genera has been insisted upon in all 

 the preceding dicotyledonous floras, not a single fruit even re- 

 motely resembling those of any one of these genera has been found 

 among them. I do not yet realize the full significance of the fact, 

 but so far most of the early dicotyledonous seeds seem to.be clustered 

 in fruit spikes as in Alnvs and Platanus and the Urticece, and few 

 seem of the nature of single-seeded nuts : nor up to this horizon is 

 there any indication of winged-seeds such as the Maple, Ash and 

 Elm, The oldest leguminous pods seem those from Reading. Next 

 in ao-e, and possibly even to some extent contemporaneous with the 

 London Clay, is the flora of Alum Bay, also met with in France in 

 the Gres du Soissonais. It is specially characterized by leaves of 



