CIVIL ARCHITECTURE. 



557 



the spot) being laid, was kept from running off by the 

 aforesaid sod or earthen dykes ; and this operation being 

 repeated, the wall was raised to the desired height. Mr A. 

 Fraser Tytler, professor of Civil History in Edinburgh, 

 afterwards Lord Woodhousclee, examined these forts 

 in 1782, and furnished an excellent paper on the sub- 

 ject, which is inserted in the Transactions of the Royal 

 Society of Edinburgh. His opinion is, that the walls 

 or mounds have been built between two rows of wood- 

 en stakes, secured together by boughs of trees, ha- 

 ving the spaces between these rows filled with stones, 

 earth, and trunks and boughs of trees, and that the 

 vitrification took place during the destruction of the 

 mass by fire. Cordiner considers each fort to be an ex- 

 tinguished volcano, and that the wells found in the in- 

 elosures are the craters ; but in Craig Phadrick there is 

 no well, and the space upon which the fort is construct- 

 ed is a mass of undisturbed pudding-stone rock, and it 

 would have been a singular circumstance, to have found 

 small volcanos so conveniently situated for a chain of 

 signals. But setting aside the theories respecting the 

 mode of construction, it is quite certain that vestiges of 

 these rude fortresses now exist, and that they must have 

 been formed by artificial means, to serve the purposes of 

 the inhabitants in a very remote period. In that view, 

 they must be allowed to be works of great magnitude 

 and labour, and they now form a singular feature in the 

 country, and a subject worthy the notice of the antiquary 

 and historian. 



The circular towers, which are numerous, appear to 

 have been sometimes placed singly, as that of Dunalis- 

 kaig, on the Dornoch Firth in Ross-shire ; sometimes one 

 large tower is found, encompassed by two or three small- 

 er ones placed about 5 or 6 paces from the large one, 

 and from each other, as Dun Dornadilla, near Tongue 

 in Sutherland ; and sometimes several large towers are 

 found situated at a considerable distance from each other 

 Sltn Elg. in the same district, as in Glen Elg in Inverness-shire. 



With regard to the latter, we have been enabled to 

 give the following account of their present state, by the 

 kindness of Mr John Mitchell, the general inspector of 

 the parliamentary roads now carrying on in the north of 

 Scotland, who has been so obliging as to examine and 

 measure them at our request. Glen Elg is the name of 

 a considerable district in the Mainland, lying opposite the 

 south-eastern part* of the isle of Skye, from which it is on- 

 ly divided by a narrow arm of the sea. It consists chiefly 

 of two valleys or glens, that is to say, Glen More or the 

 great glen, and Glen Beg or the little glen. Both of 

 these face towards the isle of Skye. Four of the round 

 towers stand in Glen Beg, and they are all situated on 

 the north-east side of its river. The first stands about a 

 mile and a half from the sea shore, upon the west shoul- 

 der of the ridge which separates the two glens : it is con- 

 siderably elevated above the bottom of the valley, and 

 commands a view of the narrow part of the strait called 

 Kyle Rhea, and also a considerable portion of the south- 

 ern end of the isle of Skye towards the island and bay of 

 Oroiisay. This tower has been totally demolished ; a 

 circle of loose stones only distinguishes its site ; they are 

 generally large, and naturally well shaped. From this a 

 second tower is seen, at the distance of about half a mile 

 from the former. It stands upon a piece of flat ground 

 close by the side of the river. It is ofa circular form, and 

 the external circumferti.ee is 178 ftet. Of this, 1 12 feet 

 appears above the mrface of the ground, only sufficient 

 to shew the thickness of the wall to be 10 feet ; but 66 

 feet of the i.orth-east quarter is still 26 feet high. This 

 part is 10 feet thick at the foundation, and 4 feet at the 



top. Near the western extreity of tke part which is 

 standing, but on the northern part of the general circle, 

 there is an entrance doorway 6 feet high, and 3 feet 6 

 inches wide. The bottom of this doorway is a little be- 

 low the present surface on the outside, but on the same 

 level with the surface within the tower. On entering, 

 this doorway, there is an entrance to a passage nearly iu 

 the middle of the wall, which is 5 feet 9 inches high, and 

 2 feet 8 inches wide at the bottom, and from 12 to 20 

 inclies at the top, which is lintelled with flat stones. 

 There is no other opening but the aforesaid doorway ia 

 this standing part. The difference between the top and 

 bottom thickness of the wall is all taken off the outside 

 by a concave batter, resembling that of the Eddyatone 

 light-house, the inside surface of the wall being nearly 

 perpendicular. Nearly in the middle of the wall, there 

 are four heights of passages over each other. The lower 

 one is ascended from the doorway, by a few steps and an 

 inclined plane to the height of about five feet. The pas- 

 sage, as has already been observed, is about 5 feet 9 

 inches high, and 2 feet 8 inches wide, and is lintelled 

 with thin flat stones. The upper passages are from 3 

 to 4 feet high, and 12 to 20 inches wide. Within the 

 tower, immediately over the doorway, there are seven 

 small openings perpendicularly over each other, and on- 

 ly separated from each other by thin flag stones, which 

 also alternately form the divisions between the passages. 

 The lowest of these openings are each about 20 inches 

 square, the uppermost 18 inches square. Near the other 

 extremity of the standing part, there is another range of 

 openings exactly similar to those already mentioned ; and 

 half way between them, there are two openings of about 

 the same size, and on a level with the uppermost of the 

 said ranges. On a level with the top of these openings, a 

 course of thin flag stones project from six to eight inches 

 all along the part of the wall which now remains. The 

 external face of the tower is composed of stonea natural- 

 ly well shaped, laid in a very close and regular manner. For 

 about six feet in height, the stones are rather of a larger 

 size than those above them, but all are equally well 

 laid. In the interior surface of the tower they are also 

 well laid, but not quite so perfectly as in the external 

 surface. In the passages the surfaces are more irregular ; 

 but in no place is there any appearance of any sort of 

 mortar, or the use of hammer, or other instrument. From 

 this secend tower there is seen, at the distance of about 

 600 yards, a third, standing in the bosom of the hill 

 about 400 yards from the river, and on ground elevated 

 about 30 feet above it. It is of a circular form ; the 

 external circumference is 165 feet. Of this, 81 feet of 

 the eastern quarter is 21 feet in height, the wall is 12" 

 feet thick at the bottom, and G at the top. The remain- 

 ing 84 feet of the circumference ij about 12 feet high oa 

 the lower side, but owing to the steep declivity of the 

 ground, the top of this part rises only to the level of the 

 surface of the ground on the upper side. There is no 

 entrance visible on the outside of what now remains. 

 Nearly in the middle of the wall, there are three stories 

 of passages divided from each other by thin flags; they 

 are each four feet in width, and from 3 to 3 feet 6 inches 

 in heighi. The bottom of the lowest passage is about 

 7 feet above the surface of the ground. 1 here is in 

 the part now standing, five openings similar to those de- 

 scribed in No. 2. They are each from 2 feet to 18 inches 

 square ; the external surface is built with a straight bat- 

 ter ; the internal nearly perpendicular ; the masonry is 

 composed of stones rather larger than the former, not of 

 BO good a quality, but quite as regularly laid. The 

 whole of the interior surface of this tower appears as if. 



