III. Stations, Castles, Watch-towers, and Roads, for the accommodation of the soldiery who manned the Barrier, and for the transmission of military stores. These lie, for the most part, between the stone wall and the earthen rampart. The whole of the works proceed from one side of the island to the other in a nearly direct line, and in comparatively close companionship. The stone wall and earthen rampart are generally within sixty or seventy yards of each other.[14] The distance between them, however, varies according to the nature of the country. Sometimes they are so close as barely to admit of the passage of the military way between them, whilst, in one or two instances, they are upwards of half-a-mile apart. It is in the high grounds of the central region that they are most widely separated. Midway between the seas, the country attains a considerable elevation; here the stone wall seeks the highest ridges, but the vallum, forsaking for a while its usual companion, runs along the adjacent valley. Both works are, however, so arranged as to afford each other the greatest amount of support which the nature of the country allows. _PLATE II._ [Illustration: PLAN OF THE BARRIER BETWEEN CILURNUM AND MAGNA AFTER HORSLEY.] [Illustration: _A PLAN OF CILURNUM AFTER WARBURTON WITH PART OF THE PLAN OF THE STONE WALL AND VALLUM._ _Shewing how they are connected at the Stations, and by their mutual relation to one another must have been one entire united Defence or Fortification._ ] _Reid Litho. 117 Pilgrim Street Newcastle_ The stone wall extends from Wall’s-end on the Tyne, to Bowness on the Solway, a space which Horsley estimates at sixty-eight miles and three furlongs—the turf wall falls short of this distance by about three miles at each end, terminating at Newcastle on the east side, and at Drumburgh on the west. [Sidenote: GENERAL VIEW OF THE WORKS.] The Map of the Wall, the more detailed Plans of several parts of it in Plate II, and the Sections given in a subsequent page, will afford a pretty correct idea of the general arrangement of the works. Most writers who have treated of the Roman remains in Britain, have considered that the two lines of fortification are the works of different periods. The earth-wall, or Vallum, has generally been ascribed to Hadrian, but the stone wall, or Murus, to Septimius Severus. This is the opinion of Horsley, whose judgment is always deserving of the highest consideration. Deferring to a subsequent period the discussion of this question, it will be convenient, meanwhile, to speak of the works as being but different parts of one great engineering scheme. [Sidenote: THE COURSE OF THE WALL.]