THEIR DISTRIBUTION. 113 



monocotyledonous orders, filices, lycopodiacese, equisetaceae, and cyca- 

 dea2, but fragments of dicotyledonouis plants also occur with them. 



The least ancient group of fossil plants, which are enclosed in strata 

 above the chalk, are a mingled suite of monocotyledonous and dioc- 

 tyledonous tribes, both terrestrial and lacustrine, bearing considerable 

 analogy to plants now in existence. The greater number of fossil shells 

 are certainly marine, but those which lie in layers amidst the monoco- 

 tyledonous plants of the carboniferous formation, belong almost wholly 

 to fresh-water genera, now in existence. Other local aggregations of 

 fresh-water shells occur in the upper part of the oolitic series of rocks ; 

 but a general deposit of this kind occurs among the most recent, and 

 contains species very similar to those that now exist. 



The greater portion of the most ancient fossil shells, &c. belong to 

 genera now extinct, as the productaj, spirifera, pentameri, orthoceratites, 

 trilobites, and many genera of crinoidea ; and on the other hand, the 

 least ancient of the fossils, though specifically distinct from existing races, 

 are mostly included in the same genera. 



But the most important results to geology, arising from the con- 

 templation of organic remains, are founded on a minute scrutiny of their 

 specific characters, and a careful register of their localities in the strata. 

 It is not enough for the rigid accuracy of modern inquiry, to say that 

 a given rock contains corals, shells, and bones of fishes ; but we must 

 know the particular species, and determine all the circumstances of 

 their occurrence. The more exact and extended our researches on this 

 subject become, the more clear will be our statements on the succession 

 of created beings, the more certain our applications of zoological prin- 

 ciples to determine the relative antiquity of rocks, and the more satis- 

 factory our views of the formation of the strata. Works which, like 

 the present, profess to describe the rocks and fossils of a particular dis- 

 trict, lose a large portion of their utility if they are composed without 

 reference to general principles. It is in such local catalogues that the 



Q 





