uch undertakings in conjunction
with Cavalry are unconditionally necessary.

The attempt to break up communications by well-mounted officers
patrols boldly pushed forward in advance would seem to offer even less
chances of securing permanent results. They, too, will find the
country obstructed by the armed population, or by troops in the act of
concentration. Even weak detachments or patrols along the railway
would suffice to effectively resist them; they can depend for success
only on their rapidity and cunning. But most rivers are unfordable,
and in the woods patrols can hardly venture, because every tree may
shelter a man with a rifle. Once they leave the roads, their pace
diminishes; they easily lose their direction; nowhere can they obtain
security for rest and food, even if they are fortunate enough in
procuring any. If, in spite of all this, they do happen to succeed in
blowing up a railway or cutting a telegraph, the effect is
infinitesimal.

The patrol itself will find its chances of escape decreasing in exact
proportion to the distance it has penetrated into the enemy's country.

The greater the number of the patrols employed, the more irreplaceable
will the inevitable loss become; for it will always be the best
officers, the men who put most energy and determination into the
execution of their instructions, who are the most likely to fall
victims to their courage and audacity.

Premature advance of the Cavalry during mobilization and concentration
can only procure information of little or no importance, for the
existing railways, the direction of the frontiers, and the peace-time
distribution of the troops reveal all this to the General Staff
beforehand. These, together with the secret service, political
conditions obtaining at the moment, and press intelligence, will
enable one to forecast with some degree of precision the general
situation.

Now, the Cavalry can hardly expect to attain more--indeed, it is
doubtful whether they would succeed even in confirming what is already
known, for the difficulties to be overcome, as we have seen, are
numerous, and nowhere can one find completed situations from which to
make reliable deductions. At most they can determine that certain
places are already occupied, and that the traffic on certain lines is
considerable, things that one knew _a priori_, which, therefore, are
not worth any serious sacrifice. Moreover, it is exceedingly doubtful
whether, at such an ear