96 M. De Candolle oti Fossil Vegetables. 



vegetation only presents gradual and limited changes. Some of 

 its species have supplied the place of other analogous ones, in a 

 manner more or less rapid, more or less complete. On the 

 contrary, the transition from one period to another period is 

 always exceedingly apparent, in every point of view : even the 

 genera are rarely the same ; the numerical proportion of the 

 classes is very different, and the species are never the same. 



These four periods correspond to four great series of forma- 

 tions, an arrangement which many geologists admit as the result 

 of entirely different considerations. 



The Jirst period, extending from the first transition, rises to 

 the termination of the coal formation, is characterized by an en- 

 ormous proportion of cryptogameae, especially by these arbo- 

 rescent varieties of ferns, equisetacese, and lycopodiaceae, of which 

 we now scarcely can find any examples, and solely in the hottest 

 climates. The ocean has covered this remarkable vegetation, 

 since in the magnesian limestone, very few species are found, and 

 these are marine. 



The second period presents a peculiar vegetation, which, as 

 yet, is little known. To the variegated sandstone, which con- 

 tains a few more phanerogameae than cryptogameas, and all of 

 which are very diflFerent from those of the first period, has 

 succeeded a long salt water inundation, as indicated by the 

 shell limestone. 



With the third period commences the reign of the cycadea^ — 

 of that anomalous family, which botanists have alternately tossed 

 from class to class, and which appears to form now a class of 

 dicotyledons very near to the cryptogameae. This family alone 

 constitutes the half of the vegetables of this period • the true 

 cryptogameag at first form only a third, then they mount up 

 to about a half of all the species ; and then again the sea ap- 

 pears to have destroyed this extraordinary vegetation. The 

 thickness of the chalk bed shows that this submersion had con- 

 tinued for many ages. 



Finally, xhe fourth period, of which our epoch forms a part, 

 is characterized by the predominance of the phanerogaraeae 

 over the cryptogameae. It would appear that one salt and 

 three fresh water inundations have four times prevailed over 

 the surface during that period, and destroyed at four different 

 times the vegetable species, previous to the appearance of those 



I 



