erosive and alluvial action of the water is permanent, that the inequalities of the slopes of the mountain are smoothed down by it to one regular slope. Hence, therefore, the idea of mountain defence would assume that, when a mountain ran about parallel with the front to be defended, it was to be regarded as a great obstacle to approach, as a kind of rampart, the gates of which were formed by the valleys. The real defence was then to be made on the crest of this rampart, (that is, on the edge of the plateau which crowned the mountain) and cut the valleys transversely. If the line of the principal mountain-chain formed somewhat of a right angle with the front of defence, then one of the principal branches would be selected to be used instead; thus the line chosen would be parallel to one of the principal valleys, and run up to the principal ridge, which might be regarded as the extremity.
We have noticed this scheme for mountain defence founded on the geological structure of the earth, because it really presented itself in theory for some time, and in the so-called “theory of ground” the laws of the process of aqueous action have been mixed up with the conduct of war.
But all this is so full of false hypotheses and incorrect substitutions, that when these are abstracted, nothing in reality remains to serve as the basis of any kind of a system.
The principal ridges of real mountains are far too impracticable and inhospitable to place large masses of troops upon them; it is often the same with the adjacent ridges, they are often too short and irregular. Plateaux do not exist on all mountain ridges, and where they are to be found they are mostly narrow, and therefore unfit to accommodate many troops; indeed, there are few mountains which, closely examined, will be found surmounted by an uninterrupted ridge, or which have their sides at such an angle that they form in some measure practicable slopes, or, at least, a succession of terraces. The principal ridge winds, bends, and splits