

INTRODUCTION. XXXI 



progress of discovery, be found, like preceding systems, to be 

 essentially erroneous. If a work of petralogy were there- 

 fore founded upon this theory, it must fell with it : and no 

 writer of judgement or industry would choose to risk his 

 labour upon such an unoertain foundation. Nay, if the 

 theory were invincible, the arrangement would still be im- 

 proper for a student of petralogy ; who must follow the best 

 mineralogical authors, and arrange substances according to 

 their chemical compositions, and other infallible rules arising 

 from the nature and appearance of the substances themselves, 

 whether they exist in nature or in cabinets. A general 

 treatise on rocks therefore cannot be founded on any theory 

 of their formation, however plausible ; as the opinions of the 

 author will be biassed by that theory, and he will be inclined, 

 like Buffon, to reject or pass in silence any substance which 

 interferes with his preconceptions. Thus jasper is totally 

 omitted by Werner, though it form a chain of mountains in 

 Siberia, of more than a thousand miles, extending even to 

 the islands between that region and America. A disciple of 

 Werner is therefore embarrassed when he sees specimens of 

 rocks, not disposed in a theoretical sequence; and, in his 

 vindication, boldly asserts that rocks can only be studied in 

 nature, where the formations indicate the series of substances. 

 But as this argument would be ridiculous if applied to litho- 

 logy, or the knowledge of small or precious stones, so it is 

 equally inapplicable to petralogy; the distinctions between 

 large stones being as certain, and still more useful to society. 

 The knowledge of small and precious stones has been accom- 

 panied and greatly assisted by the constant introduction of 

 new denominations, which at present amount to about two 

 hundred; while the rocks of Werner do not exceed sixty, 

 although the distinctions between the rocks be not only 

 more numerous, but more apparent than those between the 

 parasitical stones. If the systems of botany and zoology 

 were founded on progressive formations, it is evident that no 

 two authoi-s could agree upon the links of the chain ; and 

 such systems have accordingly been founded upon character- 



