d consists in
reconnoissances made by skillful officers and light troops; the third,
in questioning prisoners of war; the fourth, in forming hypotheses of
probabilities. This last idea I will enlarge upon farther on. There is
also a fifth method,--that of signals. Although this is used rather for
indicating the presence of the enemy than for forming conclusions as to
his designs, it may be classed with the others.

Spies will enable a general to learn more surely than by any other
agency what is going on in the midst of the enemy's camps; for
reconnoissances, however well made, can give no information of any thing
beyond the line of the advanced guard. I do not mean to say that they
should not be resorted to, for we must use every means of gaining
information; but I do say that their results are small and not to be
depended upon. Reports of prisoners are often useful, but it is
generally dangerous to credit them. A skillful chief of staff will
always be able to select intelligent officers who can so frame their
questions as to elicit important information from prisoners and
deserters.

The partisans who are sent to hang around the enemy's lines of
operations may doubtless learn something of his movements; but it is
almost impossible to communicate with them and receive the information
they possess. An extensive system of espionage will generally be
successful: it is, however, difficult for a spy to penetrate to the
general's closet and learn the secret plans he may form: it is best for
him, therefore, to limit himself to information of what he sees with his
own eyes or hears from reliable persons. Even when the general receives
from his spies information of movements, he still knows nothing of those
which may since have taken place, nor of what the enemy is going finally
to attempt. Suppose, for example, he learns that such a corps has passed
through Jena toward Weimar, and that another has passed through Gera
toward Naumburg: he must still ask himself the questions, Where are they
going, and what enterprise are they engaged in? These things the most
skillful spy cannot learn.

When armies camped in tents and in a single mass, information of the
enemy's operations was certain, because reconnoitering-parties could be
thrown forward in sight of the camps, and the spies could report
accurately their movements; but with the existing organization into
corps d'armee which either canton or bivouac, it is very difficult to