344 J. Starkie Gardner — Mesozoic Angiosperms. 



peared, in which he discards his previous artificial classification, and 

 adopts a botanical one. The Gymnospermous were separated from 

 the Angiospermous Phanerogams, the former being considered inter- 

 mediate between the Cryptogams and true Phanerogams, a proposi- 

 tion which, though placed every year on a firmer basis, is not yet 

 adequately recognized by botanists. Brongniart's system was ac- 

 cepted by almost all writers on the subject, with the notable excep- 

 tion of Lindley and Hutton, who included Cycads and Conifers 

 among Exogens or Dicotyledons, a method still followed, unfor- 

 tunately for palaeontologists, by some of the ablest botanists of the 

 day. They further included Stigmaria, Annularia and Asterophyllites 

 among true Dicotyledons, Noeggerathia is placed among the 

 Palms, and Catamites is attached to Juncus. " One of the most 

 remarkable aberrations of the book," Mr. Ward remarks, "is the 

 pertinacity with which the authors contend for the existence of 

 Cactaceous and Euphorbiaceous plants in the Coal-measures." They 

 in fact utterly ignored the principles laid down by Brongniart 

 and previous writers, and entirely dissented from the theory of 

 progressive development. Cotta's important treatise on fossil wood 

 was followed a year later, in 1833, by Witham's work on the same 

 subject, and by Zenker's 1 description of Cretaceous plants from 

 Blankenburg in the Harz district, the first attempt to treat dicoty- 

 ledonous fossils systematically. In 1840, Kossmassler proposed 2 to 

 modify Sternberg's method of classifying all fossil dicotyledonous 

 leaves as Pliyllites, substituting a compound from it and the genus to 

 which the leaves seemed to belong, as DaphnopliylUtes, etc. Thence- 

 forward the plan of naming and describing fossil as if they were 

 real plants, became definitely adopted. The terms Carpolithes, 

 Pliyllites, etc., are still employed where there is absolutely no means 

 of assigning a fossil to any definite group in the vegetable kingdom, 

 but the tendency is far greater towards unwarrantably assimilating 

 leaves with living genera which they resemble, than the employ- 

 ment of any termination, such as " ites," e.g. Pinites, to suggest that 

 the resemblance may be only apparent. 



Passing on to the question of classification, the result of palaso- 

 botanical research so far has been to necessitate certain modifications 

 in order to bring the botanical system into accord with the succes- 

 sive appearance of classes of plants in time. The grouping that best 

 meets this requirement is according to Mr. Ward : 



Cryptogams 



f Gymnosperms t Cycadacese 



< Coniferae 

 I ( Gnetacese 



^ l Monocotyledons 



qs < ( Apetalse 



( Dicotyledons < Polypetalse 

 ( Gamopetalse 



The chief modification introduced is the elevation of the Gym- 

 nosperms to the rank of a class, while the Monocotyledons and 



1 Beitrage zur Naturgeschichte der TJrwelt, Jena, 1833. 



2 Beitrage zur Versteinerungskunde, vol. i. die V. des Braunkohl. aus der Gegend 

 von Altsattel in Bbhmen. 



