ii Desmaresf s published Memoirs 61 



fresh observations drawn from extended journeys, and thus 

 making his conclusions rest on an ever- widening basis of 

 accurately determined fact. 



The Memoir, as finally published, was divided into 

 three parts, two of which appeared together, the third not 

 until three years later. In the first part, the author 

 narrated his observations in Auvergne and other districts, 

 bearing on the nature of basalt. Time would fail us were 

 we to try to follow him in his survey of the regions 

 where he found the evidence which he brought forward. 

 Let me refer merely to the concluding pages, in which he 

 states his opinion as to the origin of the columnar rock 

 which he had tracked with such diligence from district to 

 district. His account, he remarks, would be incomplete 

 if he did not indicate at the same time the materials 

 which have been melted by the fire in order to produce 

 basalt. He had collected a series of specimens of granite 

 which he believed to represent these materials. They 

 had undergone different degrees of alteration, some showing 

 still their fusible spar, quartz or other minerals, while 

 others had partly undergone complete fusion. He had 

 convinced himself that various other volcanic rocks besides 

 basalt had resulted from the fusion of granite, the base of 

 which may have been completely melted, while the quartz 

 of the original rock remained unchanged. He was not 

 aware that the difference of chemical composition demon- 

 strates that the melting of granite could never have pro- 

 duced basalt. 



These ideas, which we now know to be erroneous, might 

 readily occur to the early observers. It is undoubtedly 

 true that pieces of more or less completely melted granite 



