Naturally this view encounters opposition. Some contend that the whole triumph of the Art consists in solving both problems simultaneously. They reason that it is superfluous to seek an encounter with the enemy's Cavalry. Cavalry duels only lead to the mutual destruction of both parties. They maintain that one ought to advance, in the interests both of security and screening, on a certain breadth of front. If, then, circumstances compel one to fight, one must concentrate quickly, and after the combat gain again the necessary degree of extension to cover the front of the Army. They would leave reconnaissance to be carried out by rapidly advancing patrols, which evade those of the enemy, find cover in the ground, gain advantageous points of observation on the flanks and in rear of the opponent, thus obtaining their objects in spite of the enemy. I hold it to be a grave error of judgment to believe that any systematic application of this line of action will give sufficient results. Advantages in war must be fought for; they cannot be filched. Was it not difficult enough in 1870-1871 to obtain reliable information, although we had no true Cavalry opponent against us, and still more difficult to get that news through to Headquarters in time? How much more difficult, therefore, will it not be in the future, when we can no longer count on controlling unconditionally the country between the two Armies, and the enemy's Cavalry hold the field as well as we! Who will then guarantee that our patrols will really penetrate the enemy's screen; above all, that their reports will get back through the district controlled by the enemy's Cavalry in time enough to be of use to Headquarters in forming its decisions? If the patrols are compelled to elude those of the enemy, to seek for intersected country, and to make detours, one cannot count on the requisite rapidity; and the greater the total numbers brought together in modern War, the greater the distances become which have to be reckoned with.