96 



decide whether the rock be syenite or porphyry. The syenite 

 in these cases, however, is never, so far as I know, that variety 



• 



consisting of compact feldspar and hornblende (diorite) , which 

 occurs as a member of the overlying family of rocks, but that 

 variety composed essentially of feldspar, quartz, and horn- 

 blende." This is a concise and accurate statement of the 

 conclusion which I have reached by a re-survey of the facts. 

 The principal result of my study of these rocks is simply a 

 confirmation of the views expressed by this sagacious observer 

 more than a generation ago. 



The superposition of the petrosilex on the granite is proved 

 by the following facts : In the first place, the entire aspect of 

 the petrosilex tells us that it is the newer rock ; it is much less 

 crystalline than the granite, and far more generally stratified. 

 It is stratified as the rule, and exotic as the exception, while 

 the granite is exactly the reverse. The granite is often clearly 

 eruptive through the petrosilex ; but I have never observed an 

 indubitable instance of the opposite case. An inspection of 

 the map will suffice to show that the petrosilex is always 

 adjacent to the granite, and especially that no other rock ever 

 comes stratigraphically or geographically between these two. 



I have alluded, on a preceding page, to the very numerous 

 instances which this region affords of rocks transitional between 

 petrosilex and granite, and to the general abundance of the 

 evidence that, in this region, these two important lithologic 

 types have had a common origin. Much of this evidence 

 has been given in the foregoing descriptions of the petrosilex 

 and granite, and, I should not venture to introduce the fol- 

 lowing summary of the whole were it not that its intimate re- 

 lations to the petrosilex constitute our principal and only con- 

 clusive reason for including the granite in the Huronian system. 

 I take the petrosilex as the type of the Huronian series in Eastern 

 Massachusetts, and assume that whatever appears to be strati- 

 graphically and lithologically connected with it, as the granite 

 certainly does, must be referred to the same general horizon ! It 

 might be argued that the transition from petrosilex to granite is 



