486 



RUM. GEOLOGY. 



the weather having prevented me from ascending it; 

 yet no error will probably be committed by adopting 

 this conjecture. Under various interruptions, which will: 

 be noticed more properly when describing the trap, it 

 extends to a considerable distance in the interior of the 

 island, and at length terminates its course on the shores 

 of Harris. It is difficult to trace in it any appearance 

 of regularity as it occurs in the mountains ; although it 

 has a tendency to the same obscurely bedded disposition 

 as is observed in other rocks of the trap family. But 

 on the shore now mentioned it assumes a regularly 

 bedded form, being disposed in thin horizontal strata, 

 among which are interposed equally thin beds of a rock 

 resembling basalt in its general characters, but which a 

 narrow inspection will discover to contain augit as a 

 constituent part. There is here an evident transition 

 from this rock to an apparent basalt ; while both the 

 substances in their best marked characters pass from 

 a distinctly stratified into a massive and amorphous dis- 

 position. 



Immediately where this rock ceases a new one appears, 

 forming the remaining portion of the western moun- 

 tainous part of the island. This is a syenite similar 

 to that described as constituting a large portion of 

 Sky, as well as of Mull and St. Kilda. In general, its 

 mineral characters so perfectly resemble those of the 

 rocks now alluded to, that it is superfluous to notice 

 them. In some few places indeed it possesses a pe- 

 culiarity of character which will be better described 

 when its connexion with the neighbouring rocks is traced. 



It occupies the whole of the coast from Harris to 

 the point of Bridianoch, rising into perpendicular and 

 broken cliffs of at least 500 feet in altitude. These 

 are skirted at the foot by a low projecting terrace of 

 rock, resembling the berm in fortification, against which 

 a turbulent swell is for ever breaking. I can only con- 

 jecture that this terrace is formed of the same rock 



