upwards, and generally exhibit specimens of radiated mesotype and of analcime. The cubical zeolites (chabasite) are of rare occurrence, and the mesotype is seldom granular, and never, as far as I saw, capillary. The lamellar variety of stilbite, is occasionally found filling the intervals of approximate columns. I did not observe any zeolites in the larger and more perfect columns, but in the smaller and more irregular ones they occur, though rarely.
If we were to view the island only from the southwestern side, and at half tide, we should conclude that it has been formed of three distinct deposits, or beds of basalt. Of these the lowermost appears in some places amorphous, but it is not easy to see enough of it to judge whether it actually forms a continuous bed. It is only from the analogy of Canna, and the other basaltic islands of this sea, that we should be tempted to generalize this conclusion.
The next bed, is that which is divided into those large columns which form the most conspicuous feature of Staffa, and it varies from 30 to 50 feet in thickness. The upper one appears at a distance to be an uniform mass of amorphous basalt, but on a nearer inspection it is found to consist of small columns, laid and entangled in every possible direction, often horizontal, and generally curved. It is this bed which forms the ponderous cap (as it is called) which crowns the summit of the grand façade.
Although the great columnar bed occupies but a small portion of the whole exterior face of the island, the columnar form is perhaps predominant throughout the whole. Yet it would be equally difficult, as useless, to attempt to determine its proportion to the amorphous part where they are irregularly mixed, as they are at the northern and eastern sides. On these sides also the division into distinct beds such as I have described above, is by no means easy to trace, and possibly it does not exist.