OVERLYING AND TRAP ROCKS. 109 



evident that the boundaries of rocks so nearly allied, 

 must be evanescent. It is under this last composition 

 that greenstone so often possesses the granitic cha- 

 racter ; being often, when separated from its con- 

 nexions, undistinguishable from specimens occurring 

 in granite. In many cases, under an apparent resem- 

 blance, the existence of quartz in the latter, forms a 

 distinction, though not a certain one : while the presence 

 of sphene and other minerals is riot sufficiently con- 

 stant for this purpose. It is one of the cases of the 

 intimacy of character among unstratified rocks far 

 distant in geological time. 



I have shown that the old distinction between 

 Greenstone and SYENITE is a superficial and incorrect 

 one ; but this term must be retained for the sake of 

 the triple and quadruple compounds in this family 

 which become, rnineralogically, granites ; while I 

 have rescinded it out of that one, as I have done 

 greenstone both from granite and hornblende schist, 

 to prevent future geological confusion. If I retain it 

 for the binary compounds of hornblende, I would wil- 

 lingly limit it to those with common felspar, and in 

 which this latter predominates ; while it is plain that 

 its limits towards greenstone must become evanescent. 

 But averse to more innovation than is avoidable, it may 

 remain as the name of any binary compound also, in 

 which clinkstone or compact felspar similarly pre- 

 dominate, while equally then graduating into the 

 analogous greenstones. 



If the distinction of AUGIT ROCK is new, it is much 

 needed, as it appears to be very abundant in nature, 

 and, in my own experience, more so than greenstone; 

 with which it has been so confounded. It may be dif- 

 ficult to distinguish Augit from hornblende, when in a 

 minute state and much intermixed; but everv mine- 



