A third state of the sorus, which has a like effect to the last, and may either be initiated independently or by transition from the gradate, is the "mixed" state, where sporangia of different ages are irregularly interpolated between those already present. This is the final condition seen in the evolution of the Lepto sporangiate ferns, and is found in all the more advanced types.
From such intermediate stocks as those mentioned some six derivative phyla of advanced Leptosporangiate ferns may be dis tinguished, each centring round some well-known genus. Two of these comparison shows to have been derived from the Dicksoni aceae, viz., the Davallioid ferns centred round Davallia, and the Pteroids round Pteris. The Gymnogrammoid ferns are naturally grouped round Gymnogramme, and may probably be traced from the Osmundaceae, with Plagiogyria as a suggestive link. The Cyatheaceae probably gave rise on the other hand to the Blech noid ferns, with Blechnum as a central type ; and the Dryopte roids round Dryopteris. Lastly, a quite considerable number of genera may be traced as Dipteroid derivatives, from an ancestry suggested by Matonia and Dipteris. Thus at least six main evo lutionary sequences of advanced Leptosporangiate ferns, with more or less pronouncedly "mixed" condition of their sori, may be referred in origin to types already distinct in Palaeozoic, or certainly in Mesozoic time. Each of these will have pursued its own phyletic advance independently of the others. Comparison reveals that most of them, or in some respects all, show parallel features of advance in form, vascular structure, soral characters, and sporangia, and particularly in the reduced spore-output from each sporangium. There is thus wide evidence of independent
homoplastic, and even convergent evolution in the several phyla. In no respect is this clearer than in the distinctive feature of heterospory : for the Marsileaceae are referable in origin to a Schizaeoid source, while the Salviniaceae, whatever their actual relation, were of distinct origin from the Marsileaceae.
This brief abstract of the present position of the phyletic study of the Filicales can do no more than suggest how the matter stands to-day. Of all the Pteridophyta the ferns yield the most consecutive results. The living representatives of all the rest appear as isolated survivals, illuminated by fossil evidence, often as fragmentary and isolated as theirs. They raise as many evolu tionary questions as they solve. It is only in the Filicales that it is possible, by placing together the evidence from palaeontology and that derived from the living flora, to reconstruct a story which, however incomplete, is sufficiently consecutive to serve as a basis for evolutionary opinion. The sum of it is for the ferns, as also for other Pteridophyta, that they have held their own to the present day as a class which has made the best of their amphibious existence by help of profuse production of homosporous spores. But from the point of view of descent, they have led on directly to no further type of land vegetation. The sources of this must be sought elsewhere.