J. SfarTcie Gardner — Mesozoic Angiosperms. 345 



Dicotyledons are reduced to sub-classes. This is done because the 

 distinction between open and closed ovaries is regarded as a class 

 distinction, and Gyrnnosperms are not dicotyledonous, but have a 

 variable number of cotyledons ranging from 1 to 15. The conclusion 

 that Gyrnnosperms form a natural transition from Cryptogams to the 

 higher flowering plants has been arrived at by Sachs and others 

 quite independently of the evidence of fossils. 



Valuable statistics have been collected as to the actual numbers 

 of species described, and we find, so far as Angiosperms are con- 

 cerned, the result to be as follows : For the Carboniferous 8 or 

 0-5 %, Permian 3 or 0-9%, in the Bunter 4 or 18-2%, Keuper 1 or 

 2 - 4°/ . For the Jurassic we have 1 or 0.8 °/ in the Rhastic, 5 or 

 3-7 °/ in the Lias, 9 or 2-1 °/ in the Oolite, and inexplicably enough 

 only 1 or 0-8 °/ in the Wealden. The absence of flowering plants 

 in this formation is one of the mysteries of the Geological Record 

 which Mr. Ward does not elucidate. Vast delta deposits, such as 

 these, composed of most vai'ied kinds of sediments, spread over 

 many and wide areas, in places teeming with plant and other remains 

 of terrestrial life, would appear in every way fitted to have preserved 

 fair representative examples of the vegetation of the period ; yet, 

 though monocotyledonous flowering plants must have been far from 

 uncommon then, only a solitary one has been met with. To the close 

 of the Jurassics only monocotyledonous Angiosperms are found, 

 but with the Cretaceous Dicotyledons join them, and the per- 

 centage gradually increases until in the Laramie it reaches 37-5 °/ , 

 a figure only surpassed in the so-called Paleocene. The Eocene 

 would appear the age of Monocotyledons, judged by the 16-8 °/ , 

 though the Miocene possesses no less than 272 species. Mr. Ward's 

 tables have a statistical value, and enable us to see how far we 

 are advanced ; but it is needless to say that even the revision of 

 Mesozoic Angiosperms in the preceding pages of this Magazine has 

 profoundly modified the per-centages of the group in the Jurassic. 

 The supposed Carboniferous and Permian Monocotyledons rest on 

 wholly unsatisfactory material, and Angiosperms cannot at present 

 be traced with any approach to certainty farther back than the 

 Mesozoic period. The value of such a table is again lessened by 

 the possibilities of error in correlating the formations of the two 

 hemispheres. If, for instance, we took the Lower Cretaceous of 

 Europe, we should have a complete absence of Dicotyledons ; but if 

 we take the supposed equivalent of the Gault in America, we have 

 " many of our most familiar forms, the poplar, the birch, the beech, 

 the sycamore, and the oak ;" " the fig tree, the true laurel, the sasa- 

 fras, the persimmon, the maple, the walnut, the magnolia, and even 

 the apple and the plum." That the American correlations of the 

 Cretaceous and Eocene deposits of the Western hemisphere with 

 ours, are faulty, becomes more and more apparent, and that many of 

 the Dakotah plants actually belong to the genera quoted is still 

 more open to question. Destructive criticism will, for a long time to 

 come, do more to advance this study than any additions to species, 

 and it is sincerely to be hoped that Mr. Ward or some other com- 

 petent palasobotanist will devote himself to this side of the problem. 



