Page:Transactions of the Geological Society, 1st series, vol. 2.djvu/405

From Wikisource
Jump to navigation Jump to search
This page has been validated.
various parts of Scotland.

proaching them on any side they exhibit high columnar precipices, surmounted by grassy irregular plains, and surrounded by rocky shores of difficult access, which are whitened by a rapid and generally turbulent sea. It is easy to perceive that they are all of a trap formation, and with that general knowledge of two of them, Eilanakily and Eilan Wirrey, I was obliged to remain content, as a gale of wind with thick weather coming on soon after I landed on Gariveilan prevented me from extending my examination further, without endangering the boat and her crew. I trust that some future mineralogist, with better fortune, may complete the investigation which I was obliged to leave undone, and fill up the blanks which will be found in my brief account of this very picturesque and interesting spot.

The boat landed in a bay on the north side of Gariveilan, where a noble façade of columnar trap descends perpendicularly and without a break into the sea, rising to an elevation, as I should guess, of 200 feet. In picturesque effect, and in continuous profundity of shade, it excels even the celebrated cliffs of Staffa. The columns are however neither so regular nor so perpendicular as those of that island, but rather resemble the obscurely formed ones which occur on the shores of Ulva. A few here and there present nearly the same degree of regularity as those well-known specimens on the south side of Arthur's seat. This colonnade extends along a great part of the northern shore, plunging under the water to the eastward, while towards the west it is so elevated as to allow an examination of several strata, which lie beneath it. At the south side of the island the ground slopes so as nearly to meet the sea. It is on this south side, and at the eastern end, that these different strata come into view. The lowermost bed is a dark purple hornstone, of which the thickness cannot be determined, as nothing is

3 d 2