deni, in Northumberland and Teviotdale; the Selvogæ, in Eskdale, Annandale, and Nithsdale, on the shores of the Solway Firth: still West, the Novanta, in Galloway, Carrick, Kyle, and Cunningham; and on the North-west, above the Otadeni and Gadeni, the Damnii, in Clydesdale, Renfrew, Lenox, and Stirlingshire. These five nations were sometimes comprehended under the general name of the Mæatæ.

When Britain was formed into a regular Roman province, under the later emperors, the nations above enumerated were comprised in the five following grand divisions:-I. Britannia Prima, comprising the South-east and probably all the South-west of Britain. II. Britannia Secunda, containing Wales. III. Flavia Cæsariensis, containing probably the parts between the Thames and Humber on the East, and from the Lower Avon to the Ribble on the West; though some place it in the West of England. IV. Maxima Cæsariensis, containing the North of England, from the Humber and Ribble, to the wall of Severus. And subsequently, in the time of the Emperor Valens, A.D. 364. V. Valentia, comprehending the five Scottish tribes, already mentioned under the name of Mæatæ, lying between the walls of Antoninus and Severus, about to be described, which were built to prevent the incursions of the barbarous Scottish tribes into the Roman provinces. The first of these was built by Agricola, A.D. 79, nearly in the situation of the Rampart of Hadrian and wall of Severus, hereafter to be described: But in A.D. 81. Agricola built a line of very strong forts, advanced considerably North, from the Firth of Forth, on the East, to the Firth of Clyde, on the Western coast of Scotland. These, however, appear to have been insuf

ficient to restrain the progress of the barbarians after the departure of Agricola, A.D. 85; and in A.D. 120, the Emperor Hadrian planned and executed a much stronger and more important rampart. It began from Tunnocelum, or Boulness, on the Estuarium, Itunæ, or Solway Firth, near Luguvallium, or Carlisle, on the Western coast, and was continued almost in a direct line, to Segedenum, or Cousin's house, beyond Pons Elii, or Newcastleupon-Tyne, on the Eastern shore, being a distance of rather more than 68 English, or 74 Roman miles. It consisted of a principal agger or vallum, that is, a rampart, about 10 or 12 feet high, a ditch, on the North of this vallum, 9 feet deep and 11 feet wide, an agger 20 feet on the North side of this ditch, and an agger, without a ditch, 5 feet on the South of the principal agger, and nearly of as large dimensions. This work was garrisoned by soldiers stationed at proper intervals, in forts which had formed the first Wall of Agricola. Twenty years after this; A.D. 140, Lollius Urbicus, under the Emperor Antoninus, having re-conquered the Mæatæ, restored the second Wall of Agricola, which is commonly called the Vallum Antonini. This work consisted of a ditch about 12 feet wide, the principal wall or rampart, on the South brink of the ditch, whose foundations are 12 feet thick, but the height is unknown, and a military way on the South of this wall. There were forts, or stations, at the distance of every two miles, and smaller towers in the intervals between the forts.

But the greatest work of all was that of Severus, yet to be described. It was begun A.D. 209, and finished the next year, and was only a few yards to the North of Hadrian's Wall. This great work consisted of a ditch,

the dimensions of which are not known, except that it was in all respects larger and wider than that of Hadrian, on the South brink of which stood the wall, built of solid stone, and cemented with the strongest mortar. The height of this wall was 12 feet, besides the parapet, and its bredth 8 feet, defended at intervals by fortresses of three different kinds. Those called stationes were very strong garrisons, the least of them capable of containing 600 men, and having a town without their walls; the number of these was not less than 18, at an average distance of four miles from each other; but placed with some irregularity, according to the nature of the surrounding country and the exigency of defence. Besides these, there were in the intervals of the stations, 81 castella, at the distance of about 7 furlongs from each other. These were very strong forts, each exactly 64 feet square. Lastly, between every two castella were 4 turres, or turrets, 12 feet square, 324 in number, and 300 yards distant from each other. These were used as watch-towers, and, being within reach of each other, communications could be made with the utmost facility. For convenience of relieving guards, there was a military way, made of square stones, the whole length of the wall, on its South side, and communicating with each turret and castle; and at some distance, South of this, was another larger military way, paved also with square stones, communicating from station to station. The whole body of forces employed to garrison this stupendous work was not less than 10,000 men, 1600 of whom were cavalry and 600 mariners, at the points where the ramparts communicated with the shore.

The four principal Roman roads, Via stratæ, or paved

roads, hence called Streets, were, the Watling Street, from Dover to Chester, passing through Londinium, or London, Verulamium, St. Albans, Magiovintum, Dunstable, Lactodorum, Stony Stratford or Towcester, Manduessedum, Manceter, Etocetum, Wall, Pennocrucium, Stretton near Penkridge, to Deva or Deona, Chester. A branch of this communicated between Pennocrucium and Uriconium. Its etymology is uncertain, but it is perhaps corrupted from the name of Vitellianus into Vitellin or Watling Street. The Foss Way, derived from fossa, a ditch, extended from Totness in Devonshire, through Cirencester and Lincoln to North Britain. The Ikenild Street, probably so called from the Iceni, through whose country it ran, extended from Southampton, through York, to Newcastle-upon-Tyne. The Ermine Street, most probably derived from the Saxon Herrman, a warrior, signifying that it was a military road, extended from Menapia, St. David's, to Southampton. From these principal roads there were many minor branches.

Of the British Islands, Vectis was the Isle of Wight; the Cassiterides were the Scilly Islands, which are said to have been frequented by the Phoenicians; Mona Taciti, or the Mona described by Tacitus, in his Life of Agricola, is the Isle of Anglesea; and Mona Cæsaris the Isle of Man. Ierne, or Hibernia, was Ireland. The Hebudes mentioned by Pliny, Ptolemy, and Solinus, are now, by a slight corruption of the name, called the Hebrides.