Applying this idea to naval matters, we see why the very maximum of skill is required in our war mechanisms and war organizations, in their almost infinite variety and complexity.  The war mechanisms and war organizations of the military nations are capable of enormous results, but only when they are used with enormous skill.  There are no other instruments or organizations that need so much skill to handle them, because of the difficulties attending their use and the issues at stake.  Their development has been a process long and painful.  On no other things has so much money been spent; to perfect no other things have so many lives been sacrificed; on no other things, excepting possibly religion, have so many books been written; to no other things has the strenuous exertion of so many minds been devoted; in operating no other things has such a combination of talent and genius and power of will and spirit been employed.

A battleship is an instrument requiring skill to handle well, considered both as a mechanism and as an organization.  Its effective handling calls for skill not only on the part of the captain, but on the part of all hands.  The finest dreadnaught is ineffective if manned by an ineffective crew.  The number and complexity of the mechanisms on board are so great as to stagger the imagination; and the circumstances of modern warfare are so difficult that, as between two forces evenly matched as to material, a comparatively slight advantage in errors made will turn the scale in favor of the more skilful.  A difference in errors, for instance, in the relation of 9 to 8, under the conditions mentioned above, between two fleets having an equal number of similar ships, would give one side twice as many hits as the other in any given length of time.

In March, 1905, the writer published an essay in the Proceedings of the U. S. Naval Institute called “American Naval Policy,” in which the effect of initial superiority in gun-fire was shown in tables.  One table showed that an initial advantage of only 10 per cent secured an overwhelming victory by an accumulative effect.  Now a difference of 10 per cent in hits, under conditions in which the hits were about 10 per cent of the maximum, would mean, roughly speaking, the difference between 10 hits and 9 hits in a given length of time, or a difference between 90 misses and 91 misses; a difference in errors made of a little more than 1 per cent.

The conclusion to be drawn is too obvious to be stated.  Perhaps the conclusion is not broadly new; but possibly the idea is new that so small a difference in errors made will, under conditions of sufficient difficulty, produce such a tremendous difference in results.

Now, a division is more complex and more difficult to handle perfectly than is a battleship; a squadron more so than a division; a fleet more so than a squadron; a navy more so than a fleet.