Remarks an the Ancient Flora of the Earth. 121 



the commencement of the formation of every new kind of rock, 

 we observe, that the animal remains that were characteristic of 

 former formations almost entirely disappear. Entirely new 

 forms often distinguished but specifically, and as often differ- 

 ent in genus and family, appears suddenly, in the place of 

 the old forms. The more that the researches of naturalists 

 have enabled us to discriminate closely the organic remains of 

 a former world, the more do the early statements so often brought 

 forward vanish ; as if there were certain genera and species of 

 animals, which had escaped uninjured all the revolutions of the 

 earth's surface. It appears in reality to have been these import- 

 ant analogies that directed M. Brongniart in the estabhshment of 

 his principles, for we find him expressing it as his opinion, that no 

 common character exists among the Floras of the different pe- 

 riods. M. Brongniart at least, expressly maintains, that the 

 same species of plant does not occur in any two contiguous 

 periods, that among them every thing is different, and that we 

 cannot help believing, that an entirely new creation of mem- 

 bers of the vegetable kingdom produced under perfectly new 

 circumstances, may have displaced the older creation. 



We would certainly have every reason to feel quite satisfied 

 with these results of our author's researches, but for the circum- 

 stance that the knowledge of the plants of a former world 

 developes some facts, that, according to our judgment, do not 

 agree with his views. Before we, however, bring forward these 

 facts individually, we cannot help remarking, that our objec- 

 tions by no means apply to the general character of the Floras, 

 whose different periods of vegetation, the method M. Brongniart 

 has with so much research determined. The method first pointed 

 out by Alexander von Humboldt in his works on the geographi- 

 cal distribution of the plants on the present surface of the earth, 

 of characterising the floras of different regions by the quotietits 

 which determine the natural families by their union, has been 

 attempted in this department by M. Brongniart with equally suc- 

 cessful results. Hence he justly remarks, that although later 

 observations should add newly discovered forms to the indivi- 

 duals adduced by him, yet, in the essential characters, his floras 

 would remain with but an unimportant alteration. 



4 



