266 ON THE FORMATION OF ROCKS. (June, 



surface by lire or water, we should probably conclude, that 

 these agents were not necessarily instrumental in the for- 

 mation of those substances; and that we were totally igno- 

 rant of the process which nature may have adopted to form 

 those substances, and we should doubt whether those sub- 

 stances had not always existed in that state. Thus would 

 tlie investigation be left, until farther dissections, and the 

 analyses of similar constituted balls, had thrown more 

 light on the subject of our inquiries. 



In this manner the examination of the origin of the 

 rocks that form the external crust of our globe ought, 

 perhaps, to be conducted; beginning with those sub- 

 stances that have been formed under the immediate evi- 

 dence of our senses, and completely within the limits of 

 our observation, either by water or fire, and proceeding 

 to others having a direct resemblance, in structure, com- 

 ponent parts, or relative situation, or united by the chain 

 of positive analogy, to die same mode of formation; evi- 

 dently deriving their origin from the action of the same 

 agents of water or fire, until we come to the last crust, 

 beyond which we cannot penetrate; then we must drop 

 the thread of positive analogy, and not being able to make 

 a cut to the centre of the globe, be content with proba- 

 ble conjecture. 



At this point, where positive analogy finishes, and 

 probable conjecture begins, will be the natural line which 

 will divide the- rocks into f wo classes; the first class will 

 contain all those whose origin, either by fire or water, has 

 taken place under the evidence of our actual observation, 

 or those that can be traced by positive analogy to the 

 same origin. The second class comprising all those rocks 

 which have no positive analogy with either, yet contain- 



