INTRODUCTION. XXX111 



upon the denominations and definitions. For what hope of 

 any reconcilement of opinions, or any clear knowledge, when 

 the French persist at this moment in regarding basalt as com- 

 pact lava; while Dolomieu, the greatest of their mineral- 

 ogists, and at the same time a practical and sedulous observer 

 of volcanoes, has loudly declared that the basalt of the an- 

 cients is never a volcanic product ? 



Petralogy therefore, or the knowledge of rocks, must, like 

 the other branches of mineralogy, be studied in cabinets as 

 well as in nature 5 and in the substances themselves, not in 

 supposed theoretical positions : for if the student cannot dis- 

 tinguish a rock without these adventitious aids, which in the 

 great variety of nature will themselves often lead to false 

 conclusions, he may be pronounced as truly ignorant of the 

 subject, as he who cannot distinguish gems without being 

 informed of their countries, sites, and gangarts. And this 

 would be the more absurd as it is self-evident, as already ob- 

 served, that large substances must present more palpable and 

 more numerous distinguishing characteristics than the 

 minute. 



It must also be considered that Werner, by founding the 

 knowledge of rocks on a system of geognosy, has been led by 

 juxta-positions, and other accidental circumstances, observed 

 in the confined scale of Saxony, to diminish rather than to 

 enlarge the number of denominations j the result of which 

 practice would evidently be to obstruct the progress of the 

 science 5 and, as he is not versed in erudition, his own deno- 

 minations are sometimes unclassical, and so vague, as to give 

 no positive idea ; of which examples may be found in his 

 flinty slate, his slate porphyry, and his white-stone. Indeed 

 his new denominations in lithology being often founded on 

 colour, have been sometimes rejected*. To institute new de- 

 nominations, it is evident that erudition is necessary j and 

 this leads me to observe, that the study of preceding works 



* When he classes the gems as siliceous, instead of argillaceous, he confounds 

 them with the false gems (rock crystals, &c.), which are siliceous, 



VOL. I. C 



