THE PHYSICAL STRUCTURE OF THE EARTH. 33 



granite and certain other rocks were formed was attacked 

 by Sorby with great skill in a paper read before the 

 Geological Society in 1858. The microscopic spaces 

 in many minerals contain a liquid which does not entirely 

 fill the hollow, but leaves a small vacuum ; and Sorby 

 ingeniously pointed out that the rock must have solidi- 

 fied at least at a temperature high enough to expand 

 the liquid so as to fill the cavity. Sorby 's important 

 memoir laid the foundation of microscopic petrography, 

 which is now not only one of the most promising 

 branches of geological research, but which has been 

 successfully applied by Sorby himself, and by Maske- 

 lyne, to the study of meteorites. 



As regards the physical character of the earth, two 

 theories have been held : one, that of a fluid interior 

 covered by a thin crust ; the other, of a practically solid 

 sphere. The former is now generally considered by 

 physicists to be untenable. Though there is still much 

 difference of opinion, the prevailing feeling on the sub- 

 ject has been expressed by Professor Le Conte, who 

 says, 'the whole theory of igneous agencies which is 

 little less than the whole foundation of theoretic geology 

 must be reconstructed on the basis of a solid earth.' 



In 1837 Agassiz startled the scientific world by his 

 ' Discours sur 1'ancienne extension des Glaciers,' in 

 which, developing the observation already made by 

 Charpentier and Yenetz, that boulders had been trans- 

 ported to great distances, and that rocks far away from, 

 or high above, existing glaciers, are polished and 

 scratched by the action of ice, he boldly asserted the 

 existence of a ' glacial period,' during which Switzer- 



D 



