26 BULLETIN OK THE NATLTRA.L HISTORY SOCIETY. 



as the minerals are not well crystallized. The porphyries 

 and felsites correspond roughly to the most acid intru- 

 sives, the porphyrites to the intermediate ones, and 

 diabase to the more basic. 



Segregation. 



A curious, and for a long time unexplained fact 

 about volcanic rocks is that lavas erupted from the same 

 centre often differ markedl}^ in composition, some being 

 more basic, some more acid. Correspondingly, we find 

 with intrusive rocks that one part of the intrusion often 

 differs much from other parts. This variation follows 

 closely the rule alreadj'' noted, that the potash and soda 

 go mostly with a high silica percentage, the lime, iron 

 and magnesia with a low one. Hence, it can scarcely be 

 due to accidental differences in the original melted mass, 

 and it is believed that the cause lies in a tendency in all 

 molten magmas to split gradually into acid and basic 

 divisions, the potash and soda having a greater chemical 

 affinity for the silica than have lime, iron, or magnesia. 



The Rocks About St. John. 



There are around St. John almost all of the different 

 kinds of rocks whose origin we have discussed, and some 

 are of considerable interest. 



3Ietam Orphic Rocks. — To begin with the oldest, we 

 have the " Laurentian " series, of limestones, gneisses and 

 quartzites.^ These are very highly altered rocks, but 

 were once sediments probably very like those of the 

 present day. The quartzite was once a sand, composed 

 of loose rounded grains of quartz, but now all the spaces 

 between the grains are filled up with quartz similar to 



* The reader may observe the areas occupied by these and the following divisions 

 •of the rocks at St. John, by referring to the accompanying map. 



