in the military art, a small fortified place, environed on all sides with a moat, rampart, and parapet. Its use is to secure some high ground, or the passage of a river, to make good an advantageous post, to defend the lines and quarters of a siege, &c.

Forts are made of different figures and extents, according, cording as the ground requires. Some are fortified with battions, others with demibastions. Some again are in form of a square, others of a pentagon. A fort differs from a citadel, as this last is built to command some town.

Royal Fort, is one whose line of defence is at least 26 fathoms long.

Star Fort, is a fence or redoubt, constituted by re-entering and salient angles, having commonly from five to eight points, and the sides flanking each other.

Vitrified Forts, a very singular kind of structures found in the highlands and northern parts of Scotland, in which the walls have the appearance of being melted into a solid mass, so as to resemble the lava of a volcano, for which indeed they have been taken by several persons who have visited them.

These walls were taken notice of by Mr Williams an engineer, who wrote a treatise upon the subject, and was the first who supposed them to be works of art; other naturalists having attributed them to a volcanic origin. These works are commonly situated on the tops of small hills, commanding an extensive view of the adjacent valley or low country. The area on the summit, varying, as is supposed, according to the number of cattle the proprietor had to protect, or the dependents he was obliged to accommodate, is surrounded with a high and strong wall, of which the stones are melted, most of them entirely; while others, in which the fusion has not been so complete, are sunk in the vitrified matter in such a manner as to be quite enclosed with it; and in some places the fusion has been so perfect, that the ruins appear like masses of coarse glass. Mr Williams has not only absolutely determined the walls in question to be the works of art, but has even hazarded a conjecture as to the manner in which they were constructed, and which, according to him, was as follows. Two parallel dikes of earth or sod being raised, in the direction of the intended wall, with a space between them sufficient for its thickness, the fuel was put in, and set on fire. The stones best adapted for the purpose, called the plum-pudding stone, are everywhere to be found in the neighbourhood. These were laid on the fuel, and when melted, were kept by the frame of earth from running off; and by repeating the operation, the wall was raised to a sufficient height. This opinion of the stones being thrown in without any order, is thought to be confirmed by the circumstance of there not being anywhere a large one to be seen, nor a stone laid in any particular direction, nor one piece which has not in some degree been affected by the fire. Mr Williams mentions a fact tending to confirm his hypothesis, viz. of a brick kiln situated on the declivity of an eminence, so as to be exposed to the wind, which happening to rise briskly one time when the kiln was burning, so increased the heat, that the bricks were melted, and ran, like a lava, for a considerable way down the hill.

The opinion of Mr Williams has been embraced by several other authors; particularly Mr Freibairn and Dr Anderson, the latter having published two treatises upon these buildings in the Archæologia. In the same work, however, we meet with a paper by the Hon. Daines Barrington, in which the author expresses quite different sentiments. He observes, that Mr Williams, and the other antiquaries, who suppose the walls in question to be works of art, imagine that the reason of their being constructed in this manner was the ignorance of cement, which in these remote ages prevailed in Scotland: but with respect to this circumstance, he says, that if one side of the wall only was heated, and that to any considerable height, the matter in fusion would in all likelihood drop down to the bottom, without operating as any cement to the loose stones thrown in amongst it. This circumstance of the walls being vitrified only on one side, is indeed remarkable, and takes place in most of the forts of this kind to be met with at present; but with regard to it, Mr Barrington observes, that he himself has been twice in the Highlands of Scotland, and has found very few hills of any height which were clothed with wood; the trouble therefore of carrying it up to the top of such a mountain would be very considerable. But to this it might easily be replied, that we cannot by any means argue from the present state of the hills in the Highlands to their state in a very remote period of antiquity. At that time, it is neither impossible, nor in the least improbable, that most of the hills in Scotland were overgrown with wood; or at any rate, there undoubtedly was plenty of peat, which is still used as fuel in Scotland, and which affords such a strong heat as to be advantageously employed in melting iron, as we are informed by M. Magellan. A third particular mentioned by Mr Williams is, that these enclosures were intended as places of defence; and in support of this opinion he alleges, that there are dried wells found within most of them. But on this Mr Barrington observes, that shelter from the weather was also necessary, "upon the top of a bleak Scotch hill, whilst whisky (or a succedanum for it) would be often in greater request than the bare element of water." This objection, however, as well as the last, is evidently very frivolous; for these buildings might have roofs as well as any other; and whatever necessity there might be for whisky occasionally, water was certainly an indispensable requisite.

Mr Barrington having thus given his reasons for differing from the opinion of Mr Williams and the antiquaries just mentioned, proceeds to state his own. He tells us, that having travelled for 21 years the most mountainous circuit in Wales, he has frequently observed enclosures of dry stones, particularly a long tract in the western part of Merionethshire, called in the language of the country Duffryn, i.e. the vale. On first viewing these small enclosures made with walls of thick stones, he was at a loss to imagine how it could be worth while to construct such strong fences for so inconsiderable a piece of ground as they enclosed; but, on examining the adjacent country, he found it almost entirely covered with stones of a similar kind; and, of consequence, the smaller the space to be cleared, the less expensive would be the removal. "For the same reason," (ays he), "such dry walls are often of a great thickness, and sometimes the corners of the enclosures are filled with stones to a great width, this being the only possible means of procuring pasture." To a practice of the same kind our author would ascribe the origin of the works in question: but the objection occurs very strongly, that the walls in Scotland are vitrified, and it is not to be supposed that such trouble Vitrified would be taken with fences made in such a fortuitous manner. This objection, our author owns, would indeed be unanswerable, on the supposition that the vitrification was made on purpose to strengthen the walls of the forts; but (says he) may not the vitrification have been occasioned by volcanoes, or by what are called bloomeries? The same effect may be produced likewise on dry walls of stone by lightning passing along them. The loose stones in either case would not be rejected because they were gaily, and would be piled up in the fence of the enclosure: as the great point upon these occasions is to clear the ground, and remove the encumbering stones to the smallest distance. One of the advocates for the designed and not fortuitous vitrification, says, that the pieces he had procured did not resemble what is called lava. But every volcano is not necessarily an Etna or a Vesuvius; and consequently the matter disgorged from the crater must perpetually vary both in substance and form. Vitrified masses, larger or smaller, will likewise be produced by the same means. It may be contended, indeed, that pasture thus procured, by clearing the ground, would be more convenient at the bottom or on the sides, than on the top of the hill: but to this I answer, that in rocky countries you must get what pittance you can of foil, and often it will happen that the only detached and removeable stones are on the summit. When such enclosures have been made, they became very convenient for putting cattle into; and hence perhaps some of the wells which Mr Williams hath mentioned.

Our author concludes his dissertation on this subject by observing, that if vitrification answered the purpose of cement, it is very extraordinary that the ancient inhabitants of Scotland did not apply it to the houses or huts in which they constantly lived, but reserved this troublesome and expensive process merely for a fortification, which might not perhaps be used in half a century against an enemy. On this it is almost superfluous to observe, that in the ages of barbarity and bloodshed, in which these enclosures, whether natural or artificial, were supposed to be used as fortresses, war was so frequent, that a defence against an enemy might seem to be necessary every day, instead of once in half a century. Before we proceed further in the argument, however, it will be necessary to give some account of the situation and appearance of these forts.

According to Mr Cardonnel, the largest of them is situated on the hill of Knockfarill, to the south of the valley of Strathpeffer, two miles west from Dingwall in Rossshire. The enclosure is 120 feet long and 40 broad within the walls; strengthened on the outside with works at each end. A range of habitations seems to have been erected against, or under, the shade of the outward wall; of which those on the fourth side seem to have been higher and larger than those on the north. There are two wells in the middle, which, on being cleared out, filled with water. On the skirts of the hill to the fourth are many detached buildings; which, from the stratum of dung found on removing the ruins, appear plainly to have been used for securing the cattle. This place seems to have been anciently of consequence, and the residence of some powerful chief, from a road which leads through the hills to the north-west sea. To the east of the works are a number of vitrified ruins, extending for a considerable way along the ridge of the hill. The end next the fort seems to have joined the outer wall, and consisted either of two parallel walls, closed above, with a passage between them under cover, or a high wall broad enough to walk on. In this wall there is the vestige of a break about the middle, over which a bridge has been laid, to be drawn up or removed as occasion might require.

The fort next in consequence to that of Knockfarill is situated on the hill of Craig-Phadrick near Inverness, "which (says Mr Cardonnel) has this peculiar circumstance, that there appears to have been two vitrified walls quite round the area. The inner one seems to have been very high and strong; the outer wall but low: probably the space between was intended for securing their cattle, as there are no remains of dry-stone buildings, such as are found near the rest. Several parts of this outer wall appear quite entire, sticking to the firm bare rock, where it was first run. The area within the inner wall is near 80 paces long and 27 broad." Of this we have an account * by Alexander Frazer-Tytler, Esq. professor of civil history in the university of Edinburgh, who visited it in the year 1782. The hill itself is a small conical eminence, forming the eastern extremity of that ridge of mountains which bounds Loch Nevis on the north-west side. It is situated about a mile to the north of Inverness, and is accessible on two different quarters, viz. the west and south-east; the former affording entrance by a narrow level ridge joining the hills on Loch Nevis, and the latter by an easy ascent from the high ground above Inverness. On approaching the hill from the west, we first meet with a road cut through the rock from the bottom to the top, in most places 10 feet broad and nearly as deep; winding, for about 70 feet, with an easy serpentine direction, by which we gain an ascent over a steep rock otherwise quite inaccessible from that quarter. This road, in our author's opinion, is undoubtedly the work of art, and the vitrified matter on the top is the only thing which indicates the effect of fire; there being neither an appearance of pumice-stone, lava, nor basaltes, about the hill otherwise. There is indeed plenty of plum-pudding stone; which some have supposed to be of the nature of volcanic tufa; but this opinion is rejected by our author as erroneous. "But the circumstance (says he) which in my apprehension evinces, in the most satisfactory manner, that these appearances of the effect of fire on the summit of this hill are not the operation of nature but of art, is the regular order and disposition of those materials, the form of the ground, and the various traces of skill and contrivance which are yet discernible, though considerably defaced either by external violence or the obliterating hand of time." To investigate this matter regularly, he begins with the winding road already mentioned, and which is evidently cut through the rock for the purpose of gaining an easy ascent from the level ridge to the summit, which would otherwise have been impracticable. In ascending by this road, there appears, towards the middle, on the right hand, a small platform overhanging the passage, and inclining by a very gentle declivity to the very edge of the rock. Four enormous stones are placed upon the platform, and on the edge and extremity of it, which have evidently been guided by art into that position; it being impossible that they could have rested there, had they been rolled down from the higher parts. The obvious reason for placing them in such a position has been, that on an alarm of danger they might be projected into the path below, which could be done by the efforts of a very few men: and when this was done, the passage would be entirely obstructed, or at least rendered so difficult that it could be defended by a few against any number of assailants. Some other large stones are placed on an eminence to the left, probably with a view to block up a hollow channel, by which an enemy might have attempted to ascend. When we come to the top of the hill, a few feet below the rampart which crowns the whole, there appears an outward wall, approaching on the sides of the hill so near the upper rampart, as to have only a trench of 10 or 12 feet wide between them. This outward wall is in some places so low as to be almost level with the rock, though in other places it rises to the height of two or three feet; but even where lowest, it may be traced by a line of vitrified matter sticking fast to the rock all along, and nearly of the same breadth, which is about nine feet. The remains of this wall are strongly vitrified, except in one place on the north side, where, for about 70 yards, the rampart is formed only of dry stones and earth. At the east side, where the hill is more accessible, there is a prodigious mound of vitrified matter, extending itself to the thickness of above 40 feet. At the south-east corner, and adjoining to this immense mound, is an outwork, consisting of two semicircular vitrified walls, with a narrow pass cut through them in the middle; which appears to have been another, and perhaps the principal entry to the fort.

The inner wall, surrounding the summit of the hill, encloses an oblong level area of about 75 yards long and 30 broad, rounded at each of the ends like the outward wall. It is of considerable height, and nearly of the same thickness with the outward one.—It has some appearance of having been defended with four turrets or bastions: but the traces are so imperfect, that Mr Tytler does not lay much stress on his observations in this respect; a number of small tumuli of earth, with a stone in the centre, were more discernible. On the east side a portion of the internal space appears separated from the rest by two ranges of stones fixed strongly in the earth, and forming a right-angled parallelogram. "This separation (says our author) is immediately discernible by the eve, from this circumstance, that the whole of the enclosed summit has been most carefully cleared from stones, of which there is not one to be seen, unless those that form this division, and the single one in the middle of the circle of tumuli above mentioned. What has been the design of this separated space, it is difficult to conjecture. It might perhaps have marked the residence of those of a higher rank, or served as a temple for the purposes of devotion." On the east end of the large area on the summit is a well of about fix feet in diameter, which has probably been sunk very deep in the rock, though now it is filled up with rubbish to within a yard of the top.

The other fortified hills mentioned by Mr Cardonnel are those of Dun-Evan in the shire of Nairn; Tor-dun castle, near Fort Augustus; and another on the west side of Glenveas in Lochaber, three miles to the south of Fort William. The Castle hill of Finhaven, in the county of Angus, has likewise some considerable ruins of the same kind.

Dun-Evan and the hill of Finhaven have likewise been visited by Mr Tytler, who gives an account of them in the paper already quoted; of which the following is an abstract. "On the summit of the hill of Dun-Evan, whose name implies that it had been originally a place of defence, are the remains of two walls surrounding an oblong space like that of Craig Phadrick already described, but somewhat smaller in size. [Mr Cardonnel says that it is about 70 paces long and 30 broad]. There are likewise the traces of a well in the enclosed area; and at the east end are the remains of a prodigious mass of building, much more extensive than that on Craig Phadrick." Here, however, our author could not perceive any marks of fire; and Mr Williams owns that the vitrified ruins here are more waited than on Knockfarriil or Craig Phadrick. But with regard to the vitrifications here, our author is inclined to suppose Mr Williams to have been entirely in a mistake. On the Castle hill of Finhaven, however, the vitrified remains are very visible all round the summit, which is cleared of stones and levelled, unless at one end, where there is a great hollow space separated from the rest of the area, and probably destined exclusively for the keeping of cattle. The enclosed area is about 140 yards long, and upwards of 40 broad.

Besides these fortifications, the hill of Noth affords a remarkable appearance of the same kind: of which Mr Cordner gives the following description, not from his own observation, but those of a gentleman of credit who visited the place. "On the top of the hill there is an oblong hollow, as I could guess, of about an English acre, covered with a fine layer of grats: in the middle toward the east end of this hollow is a large and deep well. The hollow is surrounded on all sides with a thick rampart of stones. On three sides of this rampart, from 8 to 12 feet thick, is one compact body of stones and minerals which have been in a state of fusion, resembling a mixture of stone and iron-ore, all vitrified, calcined, and incorporated. On the north side the rampart consists of broken pieces of rock, which have the appearance of having been torn to pieces by some extraordinary violence. If the calcined compact wall exists under them, it is not at present visible."

Such are the descriptions of the most remarkable of these curious fortifications, which of late seem to have engaged the attention of the learned in a considerable degree. We have already taken notice, that by some they are supposed to be the works of art, by others the productions of a volcano. Mr Cardonnel adopts the opinion of Mr Williams as the most probable, both with respect to their use and manner of construction. Mr Tytler takes notice of the remarkable difference of opinion among those who have viewed the places in question. "It is curious to remark (says he) how the same appearances, to different observers, lead to the most opposite opinions and conclusions. The two gentlemen above mentioned (Mr Williams and Dr Anderson) seem not to have entertained the small- Vitrified Forts.

doubt, that the vitrified materials on the tops of these hills were the vestiges of works of art, and the remains of structures reared for the purposes of security and defence. The bishop of Derry, when on a tour to the north of Scotland, visited the hill of Craig Phadrick near Inverness, and expressed his opinion, that the mounds of vitrified matter were not the remains of any artificial work, but the traces of an ancient volcano. In the Phil. Trans. of the Royal Society of London for 1777, Part II. No. 25. is an account of Creek Fetterick, there termed a Volcanic hill near Inverness, in a letter from Thomas Weit, Esq. to Mr Law, F. R. S. in which the writer does not hesitate to pronounce this hill an extinguished volcano; and having sent specimens of the burnt matter for the inspection of the Royal Society, the secretary subjoins a note to the paper, intimating, that these specimens having been examined by some of the members well acquainted with volcanic productions, were by them judged to be real lava. Such was likewise the opinion of the late Andrew Crofie, Esq. who, in an account which he gave to the Philosophical Society of Edinburgh in 1789, offered some very curious conjectures with regard to the process of nature, by which he supposed the whole of this hill to have been thrown up from the bottom of the sea by the operation of intense fire.

Mr Tytler agrees with those who think the vitrified structures to be artificial works: but he differs from Mr Williams and others, who think that they were vitrified on purpose for cementing the materials together. His reason for this is, that the number of forts that show marks of vitrification, is considerable when compared with those that do not. He therefore considers the vitrification as accidental; and that it must have been accomplished in the following manner. In the rude state in which we must suppose Scotland to have been in early times, it is very probable that their buildings, both for habitation and defence, would be frequently constructed of loose stones of an irregular shape; of which, by themselves, it would scarce be possible to fabricate a wall of any tolerable strength. Hence it became necessary to use wood as well as stone in their construction. This kind of building, then, in our author's opinion, was begun by raising a double row of palliades or strong stakes in the form of the intended structure, in the same way as in that ancient mode of building described by Palladio under the name of riet-pinita à coffer, or coffer-work. These stakes were probably warped across by boughs of trees laid very closely together, so as to form two fences running parallel to each other at the distance of some feet, and so close as to confine all the materials of whatever size that were thrown in between them. Into this intermediate space Mr. Tytler supposes were thrown boughs and trunks of trees, earth and stones of all sizes, large or small as they could quarry or collect them. Very little care would be necessary in the disposition of these materials, as the outward fence would keep the mound in form. In this way it is easy to conceive that a very strong bulwark might be reared with great dispatch; which, joined to the natural advantage of a very inaccessible situation, and that improved by artful contrivances for increasing the difficulty of access, would form a structure capable of answering every purpose of security or defence. The most formidable attack against such a building would be fire, which would no doubt be always attempted, and often with success, by an enemy who undertook the siege. If the besiegers prevailed in gaining an approach to the ramparts, and, surrounding the external wall, set fire to it in several places, the conflagration must speedily have become universal, and the effect may be easily imagined. If there happened to be any wind at the time to increase the heat, the stony parts could not fail to come into fusion; and as the wood burnt away, sinking by their own weight into a solid mass, there would remain a wreck of vitrified matter tracking the spot where the ancient rampart had stood; irregular, and of unequal height, from the fortuitous and unequal distribution of the stony materials of which it had been composed. This conjecture appears very probable from their appearance at this day. They do not seem to have ever been much higher than they are at present, as the fragments that have fallen from them, even where the wall is lowest, are very inconsiderable. The durable nature of the materials would prevent them from suffering any changes by time; though from the gradual increase of the soil, they must in some places have lost considerably of their apparent height, and in others been quite covered. Mr Williams, in making a cut through the ramparts at Knockfarriil, found in many places the vitrified matter covered with peat moss half a foot thick.

In confirmation of this opinion, our author likewise urges that in the fortification on Craig Phadrick, a large portion of the outward rampart bears no marks of vitrification. The reason of this seems to be, that the steepness of the hill on that side renders a low fence of stones and turf sufficient; and no wood had probably been employed in its construction. "It appears therefore highly probable (concludes our author), that the effect of fire upon these hill fortifications has been entirely accidental; or to speak more properly, that fire has been employed not in the construction, but towards the demolition of such buildings: and for the latter purpose it would certainly prove much more efficacious than for the former." It is much to be doubted, whether it would be at all possible, even in the present day, by the utmost combination of labour and of skill, to surround a large space of ground with a double rampart of stones compacted by fire, of such height and solidity as to answer any purpose of security or defence against an enemy. Any structure of this kind must have been irregular, low, fragile, easily sealed, and quite infecere; a much weaker rampart, in short, than a simple wall of turf or wooden palliade. The vestiges yet remaining, as I have already observed, give no room to suppose that the vitrified mound has ever been much more entire than it is at present. The effect of fire upon structures reared in the manner I have supposed them to have been, will account most perfectly for their present appearance. It was from necessity that the builders of these fortifications betook themselves to a mode of structure so liable to be destroyed by fire. In those parts where stones could be easily quarried, of such size and form as to rear a rampart by themselves of sufficient strength and solidity, there was no occasion to employ wood or turf in its construction; and it was therefore proof against all af- Vitrified faults by fire. Such are the ramparts which appear on the hill of Dun-Jardel, Dun-Evan, and many others, on which there is not the smallest appearance of vitrification. But on Craig Phadrick, and the other hills, above described; where, from the nature of the rock, the stones could be procured only in irregular and generally small fragments, it was necessary to employ some such mode of construction as I have supposed; and these ramparts, though solid and well calculated for defence against every attack by force or stratagem, were not proof against an assault by fire."

Mr Cordiner is of opinion, that the vitrifications in question cannot have been the works of art, and ridicules the contrary hypothesis; though without adducing any argument against it. The hill of Noth is by him supposed to have been a volcano. He describes it as "a most majestic mountain, in general brown, with moss and heath, interperced with bare rock, in many places crumbling down. The highest part of it is a circular hill, whose verdure, as well as height, distinguishes it from the rest of the mountain. This is called the Top of Noth; and bears the strongest resemblance to every description of a volcanic mount. At the distance of many miles, one can distinguish those ridges which are the boundaries of the crater, indicating the hollow in the top." The gentleman from whom Mr Cordiner received the account of the vitrifications on the summit, informs us, that on first seeing specimens of them, he imagined that they had been pieces of stone calcined by the burning down of a castle; as he had found something very like them on the castle-hill at Cullen, in parts where the fward of grafs was broken; but on reaching the top, and viewing the appearances on it already described, he altered his opinion. "That men hardly befit (says he) might climb up with some provisions to this as a place of refuge, is probable: but that, on a barren mountain top, far from cultivated ground, half a day's journey from the plain; that there, in any period of society, man should have been tempted to build that amazing rampart, is not to be imagined: they have found it a natural and extensive fortress, and in critical circumstances have made use of it accordingly. That it has been occupied as a place of strength and of refuge, is very evident; for, some hundred yards lower down on the hill, there are the remains of another rampart or wall, consisting of loose stones piled together without any cement, carried quite round the hill. This last has been built for an additional defence to those who made their abode on the top. The top of Noth, for two-thirds downwards, is covered with a green fward; below that, it is brown with heath: this is the very reverse of the adjacent mountains; and the greater verdure of the upper part I imputed to a new soil created by the ashes of the volcano. The opening, called a well, I suppose to have been the latest crater. About a mile south, down towards the lower grounds of the Cabrock, there is a very pretty regular green hill, which I ascribe to a later eruption than those which may have formed the contiguous hills now covered with heath. There is an extraordinary luxuriant spring of water rushes out at once from the side of the hill of Noth; which is likewise some confirmation of the opinion that a volcano has some time existed there which has occasioned great hollows and reservoirs of water in the heart of the mountain. And the wild irregularities of nature through all the Cabrock, the hideous and strange projection of rocks from the sides of the hills, would seem to indicate some vast convolutions which the earth must have suffered in these parts.

"The traces of ancient volcanoes (says Mr Cordiner) are far from being unfrequent in Scotland. The hill of Finhaven is one instance; and not only abundant in this species of lava, but with tarres, or the pulvis puteolanus, an amalgame, as Condamine calls it, of calcined stones mixed with leorias and iron rust reduced to powder. The hill of Beregonium, near Dunfafnage callie, is another yielding vast quantities of pumice or leoria of different kinds; many of which are of the same species with those of the volcanic Iceland. The noble assemblage of basaltic columns at Staffa, those in the ille of Sky, and the rock Humble, are but so many evidences of the ancient volcanoes of this country. And finally, the immense stratum of pumex vitreus or Iceland agate, on the hill of Duan-hinn in Arran, is the last proof I shall bring in support of that question."

On this dispute we can only observe, that whatever side we embrace, the difficulties seem to be very great, nay almost insurmountable. When we consider the great thickness of the walls on the top of Noth, from 8 to 12 feet, and the vast mound of vitrified matter, no less than 40 feet in breadth, mentioned by Mr Tytler, we can scarce conceive it possible that less than a volcanic fire could be able to form them. We may safely allow, that, in the way this gentleman mentions, there might be considerable vitrifications formed; but that such immense masses should be brought into perfect fusion by the small quantity of fuel which could be put round them in palliades, or intermixed with the materials themselves, will be incredible to every one acquainted with the extreme difficulty with which stones of any magnitude are brought into complete fusion. We see even in the infides of furnaces, though sometimes built of no more infusible materials than common brick, no such effects follow. There is a slight vitrification indeed, but it scarcely ever penetrates to the depth of an inch or two, though very violent fires are kept up for a much longer time than we could suppose the wood surrounding those walls to require for its being consumed. In conflagrations, where houses are consumed, which are the only similar examples we have, no such effect is perceived. Even in the great fire at London in 1666, where so many buildings were destroyed, we do not hear of their walls being vitrified, though the materials of many of them were undoubtedly as fusible as the rocks and stones of Craig Phadrick, or the Top of Noth. If, on the other hand, we reject this, and adhere to the volcanic hypothesis, our difficulties are equally great. For where shall we find, in any other part of the world, an example of volcanoes ejecting lava in the form of walls enclosing a regular area? This would be attributing such a singularity to the volcanoes of Scotland as the most extravagant imagination cannot admit. We must therefore conclude, that though these ruins are certainly the works of art, we have not yet sufficient data to decide the question with respect to their construction, but that the subject requires a farther investigation. In the paper already quoted, Mr Tytler observes, that "these ancient fortifications present a more curious and interesting object of speculation, than those uncertain and indeed fruitless conjectures as to the mode in which they have been reared." This, he justly observes, must have been before the use of mortar was known; for as the country abounded in limestone, and the builders certainly would exert all their powers in giving them a proper degree of strength, it would undoubtedly have been used. Hence we are led to ascribe to these a very considerable degree of antiquity; for as the Britons were taught the use of mortar by the Romans, it is probable that we must date the origin of the structures in question before the time of the invasion of that people, or at least soon after it; so that we must look upon them to be more than 1650 years old; but how far beyond that period we are to search for their origin, does not appear. "All that we can conclude with certainty (says our author) is, that they belong to a period of extreme barbarism. They must have been constructed by a people scarcely removed from the state of savages, who lived under no impression of fixed or regulated property in land; whose only appropriated goods were their cattle; and whose sole security, in a life of constant depredation, was the retreat to the summits of those hills of difficult ascent, which they had fortified in the best manner they could. As the space enclosed was incapable of containing a great number of men, especially if occupied in part by cattle, it is presumed, that these retreats were formed chiefly for the security of the women and children of the canton, and of their herds. They could be defended by a few men, while the rest of the tribe were engaged with their enemies in the field."

Our author concludes his dissertation with a conjecture, that the forts in question were constructed, not only before the Roman invasion, but before the introduction of the rites of the Druids into Britain.