The last point concerns the other geographical properties of the points to be chosen. That fortresses on the sea, on streams and great rivers, and in mountains, are doubly effective, has been already stated to be one of the principal considerations; but there are a number of other points in connection with fortresses to which regard must be paid.
If a fortress cannot lie on the river itself, it is better not to place it near, but at a distance of ten or twelve miles from it; otherwise, the river intersects, and lowers the value of the sphere of action of the fortress in all those points above mentioned.(*)
This is not the same in mountains, because there the movement of large or small masses upon particular points is not restricted in the same degree as it is by a river. But fortresses on the enemy’s side of a mountain are not well placed, because they are difficult to succour. If they are on our side, the difficulty of laying siege to them is very great, as the mountains cut across the enemy’s line of communication. We give Olmütz, 1758, as an example.
It is easily seen that impassable forests and marshes have a similar effect to that of rivers.
The question has been often raised as to whether towns situated in a very difficult country are well or ill suited for fortresses. As they can be fortified and defended at a small expense, or be made much stronger, often impregnable, at an equal expenditure, and the services of a fortress are always more passive than active, it does not seem necessary to attach much importance to the objection that they can easily be blockaded.
If we now, in conclusion, cast a retrospective glance over our simple system of fortification for a country, we may assert that it rests on