ROMAN REMAINS IN BRITAIN.

Two Varieties of Masonry—Dover Castle—Richborough Castle—Newport Gate, Lincoln—Hadrian's Wall—Its Direction and Construction—Outworks—Ornamental Detail—Roman Roads and Camps.

The remains of Roman architecture in Britain, though numerous, do not exhibit any perfect buildings, and the workmanship in general is not equal to that of the Continental remains. The buildings seem to have been inferior and of smaller dimensions, and there is very little of ornamental detail to be found, except the tesselated pavements, of which many fine examples yet remain in the Roman villas which have been discovered from time to time in various parts of the kingdom.

The principal places where Roman remains are now to be seen are Lincoln, Dover Castle, St. Albans, Richborough Castle, Porchester, York, Cirencester, Leicester, and Colchester. But in all these there is little ornamentation or detail left, the remains consisting chiefly of plain walls, the masonry of which has peculiarities of character which mark its date. Of the masonry there are two principal varieties; the first, and that which is most readily recognised, consists of alternate layers or bands of pebbles, or small stones embedded in mortar, and tiles or flat stones. These bands consisted of three or four courses of tiles or stones laid through the wall, and were placed at two or three feet from each other, the intermediate spaces being raised with a sort of cement composed of mortar and pebbles, or sometimes rag-stones, or such materials as the country afforded. In this manner are built the Mint wall at Lincoln, the Jewry wall at Leicester, and the walls at Verulam (St. Albans), Porchester, Richborough, York, Pevensey, Chesterford, Colchester, Wroxeter, and Silchester.

The other variety consists of walls formed of square stones or ashlar, as the Roman wall in Northumberland. These are sometimes very large, as in the north gate (or Newport gate), Lincoln. Smaller kinds of ashlar, of almost cubical blocks, occur in the multangular tower and other buildings at York. The mortar used in all these walls is in general mixed with pounded brick.

It will not be necessary here to go into a description of all these buildings, but a few of the most remarkable may be mentioned; one of the most curious and interesting of these is the Pharos in the Castle of Dover, though it has undergone much alteration, particularly in the fifteenth century. "Wherever the outer casing is worn away, or has been removed by violence, the walls exhibit the usual mode of Roman building with the material of the districts; in this case with tufa or stalactite, brought perhaps from the opposite coast of France, and flint, with layers of large flat Roman bricks, some of them two feet long, each layer two courses deep, placed regularly and horizontally in the walls at equal intervals, or nearly so. No less than eight of these layers of brickwork are visible on the south-east side; other layers are apparently concealed by the external and subsequent casing of flint and stone, and where the casing of flint is perfect, quoins of stone appear at the angles. This tower is externally octagonal in form; internally the space enclosed forms a square. The doorway, recently blocked, is on the south side, and the arch, turned and faced with a single row of large Roman bricks, springs from a kind of rude impost moulding, somewhat resembling that of the Roman gateway at Lincoln; but this is not now visible. In the interior, the constructive features of the original Roman work were, before the entrance was closed up, far more visible and perfect than on the exterior, and the facing of the bricks was quite smooth; yet the effect of the alterations is here also plainly apparent, and the original windows, the arches of which are turned with Roman brick, have been filled up with flint masonry. Both the external as well as the internal facings of the entrance doorway on the south side were, a few years back, when the interior could be readily examined, far from perfect. Over this doorway were two windows, one above the other, each arched with brickwork. On the east side of the tower is a rather lofty arch faced with stone, the soffit of which, however, appears to have been turned with brick; this probably communicated with some building adjoining. Over this arch is a window now blocked up."

Richborough Castle, in Kent, is another of the most important of the Roman remains in England. It is a large parallelogram, including within it an area of five acres. The walls to the height of six feet are more than eleven feet thick, and above that ten feet eight inches; and the masonry is thus described by the Rev. C. H. Hartshorne:—"At Richborough, commencing at the ground, there are on the north side, where the masonry is displayed in its most perfect state, first of all, four courses of flint in their natural form, then three courses dressed; to these succeed two courses of binding tile, and then they rise above each other in the following order: seven courses of ashlar and two of tile; seven courses of ashlar and two of tile; seven courses of ashlar and two of tile; again, seven courses of ashlar and two of tile; eight courses of ashlar and two of tile; nine courses of ashlar. The extreme height of this wall is twenty-three feet two inches, and its thickness ten feet eight inches."

One of the most perfect and most interesting of Roman remains is the archway at Lincoln, called Newport Gate, and styled by Dr. Stukely "the noblest remnant of this sort in Britain." It was the north gate of the Roman city of Lindum, and from it a military way, called the High Street, leading to Winteringham, on the Humber, may now be traced. This still forms the principal entrance into the city from the north. It is supposed to have had a large central arch, and two smaller ones at the sides, that on the west having been destroyed, the larger being about fifteen feet, and the lesser ones seven feet in width. It is built of squared stone, out as far as the top of the arch, of remarkably large size. It is without ornament of any kind, but is said by Rickman to have had architrave and impost mouldings. That of the architrave, if it ever existed, has entirely disappeared; but there is, or was lately, a small portion of the impost moulding remaining, on the west side of the large arch. The masonry, which exhibits none of the usual binds of tiles so frequent in other buildings, will be best understood by reference to the engraving on page 21.

There is another piece of Roman work in the neighbourhood of Newport Gate, which is a piece of wall built with ashlar and binding courses of tile. It is known as the Mint Wall.

But perhaps the most interesting of all the Roman remains in Britain is the Roman Wall, which reaches across the narrow part of the island in Northumberland and Cumberland, commencing at Wallsend, on the Tyne, running through Newcastle and Carlisle, and terminating at Bowness, in Cumberland. A most interesting and fully illustrated account of this wall has been given to the world by the Rev. J. Collingwood Bruce, from whose work we have (by permission) copied the two illustrations on p. 22.

Arguments have been brought forward by some antiquaries to show that the wall and the vallum by which it is accompanied belong to two different periods; but Dr. Bruce contends that they are both to be considered as forming part of the great engineering work of Hadrian. It consists of a stone wall, or murus, and a wall of earth, or vallum. These two run always near together, but not always parallel. The vallum is likewise rather the shorter, terminating at Newcastle on the east, and at Drumburgh, about three miles from Bowness, the western extremity of the wall.

"The most striking feature in the plan, both of the murus and the vallum, is the determinate manner in which they pursue their straightforward course. The vallum makes fewer deviations from a right line than the stone wall; but as the wall traverses higher ground, this remarkable tendency is more easily detected in it than in the other. Shooting over the country, in its onward course, it only swerves from a straight line to take in its route the boldest elevations. So far from declining a hill, it uniformly selects one.

"For nineteen miles out of Newcastle the road to Carlisle runs upon the foundations of the wall, and during the summer months its dusty surface contrasts well with the surrounding verdure. Often will the traveller, after attaining some of the steep acclivities of his path, observe the road stretching for miles in an undeviating course to the east and the west of him, resembling, as Hutton expresses it, a white ribbon on a green ground. But if it never moves from a right line, except to occupy the highest points, it never fails to seize them as they occur, no matter how often it is compelled, with this view, to change its direction. It never bends in a curve, but always at an angle. Hence, along the craggy precipices between Sewingshields and Thirlwall, it is obliged to pursue a remarkably zigzag course; for it takes in its range, with the utmost pertinacity, every projecting rock."

Though no part of the wall now retains its full height, it has been calculated that when entire it was about eighteen or nineteen feet, including the parapet. Its general thickness is about eight feet, though it varies in different parts from six feet to ten and a half feet. It is "throughout the whole of its length accompanied on its northern margin by a broad and deep fosse, which, by increasing the comparative height of the wall, added greatly to its strength. This portion of the barrier may yet be traced, with trifling interruptions, from sea to sea."

The masonry of the wall is somewhat peculiar; it has none of the binding courses of tile which are, in many parts of England and on the Continent, so characteristic of Roman work. No tiles are used in the construction. The outer face of the wall on both sides is formed of squared blocks of stone, usually called ashlar, and the interior of rubble, embedded in mortar. These blocks are about eight or nine inches thick, and ten or eleven wide; their length is considerably more, sometimes as much as twenty-two inches, and tapering to the opposite end, which was firmly bedded into the rubble. The whole rested on a course of large foundation stones.

On or near the wall were placed, at tolerably regular intervals, stationary camps, or "stations," about seventeen or eighteen in number; and at still shorter intervals, that is, about a Roman mile from each other, were placed smaller towers, called, from this circumstance, "mile castles." These are in general placed against the south side of the wall, and had mostly only one entrance, which was from the south; but in the most perfect of those at Gawfields there are two entrances, one on the south, and another through the main wall on the north.

More of ornamental detail seems to have been bestowed on the architecture of these stations and mile castles than on the wall, which was intended for defence. The walls have moulded basements and cornices, of which there are woodcuts on this page and the next; the one from Vendalana (Chesterholm) exhibits also the peculiar ornamentation of the surface of the stone work, which is produced by cutting lines in various directions, either lozenge-wise or parallel, horizontal, upright, oblique, or zigzag-wise, thus producing considerable variety. In the extremely interesting Saxon crypt at Hexham, which was built out of the ruins of the Roman Wall, many varieties of this peculiar tooling, or "broaching," occur, along with ornamental mouldings, &c., and inscribed slabs, one of which has been cut to form the semicircular head of a doorway. The beautiful fragment of a capital also given was found in the station of Cilurnum (now Walwick Chesters). It has probably belonged to the portico of a temple. It appears to be a late variety of Corinthian or composite. It serves to show that there must have been considerable expense bestowed on these stations, which were, in fact, military cities, in which the commanders resided. The doorway from the station of Bird-Oswald is valuable as showing a peculiar form of door-head, cut out of a solid stone. It forms the entrance to the guard-chamber from the gateway of the station.

The Roman altars, sculptured fragments, inscribed stones, coins, implements of war, articles of personal ornament, and utensils for domestic use, which have been found along the line of the wall, are extremely numerous. But far more striking memorials of Agricola and his great successors in Britain are the Roman roads. Easy means of communication were, of course, a necessity for the Romans, dwelling, as they did, as a military garrison among a people notorious for their propensity to break into wild rebellion at a moment's notice; and hence the country was traversed by a complete system of roads leading from station to station. The method of their construction varied, but they were invariably raised above the surface of the country, and ran in an almost straight line regardless of hill and valley. The more important roads were very elaborately constructed with a foundation of hard earth, a bed of large stones, sometimes two more layers of stones and mortar, and above all the causeway paved with stones. The four most important roads were Watling Street, the Foss, Icknield Street, and Ermine Street. Of these, Watling Street ran from London to Wroxeter (Uriconium), and thence was continued into Wales, while part of the same system connected London with Dover. The Foss ran from the sea-coast, at Seaton in Devonshire, to Lincoln, with a continuation known as the High Street to the Humber. The Icknield way started from near Bury and ran to Wantage, and thence to Cirencester and Gloucester. Ermine Street ran through the Fens from London to Lincoln. These by no means exhaust the Roman roads, traces of which are to be found in almost every neighbourhood of England, but they are the "four Roman roads" so frequently mentioned in the legislation of the Middle Ages.

Equally numerous are the remains of Roman camps, constructed with great engineering skill. Even when it was necessary to remain stationary for a very brief period, the Romans were accustomed to surround the space to be occupied by the soldiers' tents by an earthen rampart with stakes at the top(agger or vallum), which was in turn surrounded by a fosse or trench (fossa), usually nine feet deep and twelve broad. The spot selected was always one that commended itself from its defensive capacities, and therefore could not be overlooked, and had a command of water. The streets were sometimes as much as a hundred feet broad, with a public meeting-place or forum near the general's tent, which was usually pitched on the highest ground. There was a vacant space of two hundred feet between the tents and the ramparts called the intervallum. The shape of the camp in later times varied according to the nature of the ground, although in the days of the Roman republic it was as a rule rectangular. Of these temporary camps the most perfect is that situated near Kirkboddo, five miles to the south-east of Forfar. It was probably constructed by Agricola; all its six gates exist, and the entrenchment, even now, seems to have lost but little of its original height. It is about two thousand two hundred and eighty feet in length, and one thousand and eighty in breadth; and, apparently because it was necessary to find lodging for more men than the camp was originally intended to hold, there is a procestrium or enclosure without the south-east angle of about one hundred thousand square feet. Permanent camps, which were smaller than the temporary camps, soon lost their original features and grew into towns.

The establishment of an infirmary (valetudinarium), a farriery (veterinarium), and a forge (fabrica) within the rampart were quickly followed by the settlement of a civilian population, and the birth of trades and industries. In many of the English towns, which by the termination cester or chester or the prefix caer betray their Roman origin, hardly a trace of the original Roman camp is to be found, but during the period of the Roman occupation of Britain the military element in them was probably in the ascendant.