56 GENERAL OUTLINE? ANt' 



storphin hills. But I need not accumulate more ex- 

 amples in this rock: enough have been adduced to 

 show that no combination of the experience of the 

 most practised geologist with that accurate eye for 

 form and character which distinguishes the painter, 

 will exempt the observer from the duty of a careful 

 manual examination where granite is concerned. 



It has been so often said that the trap rocks are cha- 

 racterized by the scalar outline from which their Swedish 

 name, now adopted by us, has been derived, that it is 

 necessary, for the sake of the geological student, to 

 examine into the truth of this assertion. That outline 

 does unquestionably occur ; but it is limited to those 

 examples where these rocks exist in the form of beds, 

 either horizontal or nearly so ; as in the little Cum- 

 bray, and in some parts of Sky, of Mull, and of the 

 neighbouring islands. But as this peculiar outline is 

 produced by the successive and unequal loss of por- 

 tions of such beds, it is evident that it may occur 

 in any stratified rock disposed in a similar manner ; 

 provided its fracture is in a direction nearly vertical to 

 the strata. It will therefore probably be found in 

 horizontal sandstones ; although at this moment no 

 very well marked instance occurs to my recollection. 

 But a thousand instances may be quoted, where the 

 trap rocks deviate from this outline; while the several 

 picturesque characters which they exhibit are so infi- 

 nitely varied, that no experience and no eye, are 

 capable of pronouncing on their nature from a distant 

 view. The marked granitic character of the Hyper- 

 sthene rock of Sky was already noticed ; and, with 

 respect to the syenites, porphyries, and claystoncs, 

 which form the interior hills of that island, they are 

 imdistinguialuible from the granite which constitutes 



