The actual practice of the British army proves that a recruit can be fully trained and be made fit in every way to take his place in his company by a six months' training, but in my opinion that is not sufficient preparation for war. The recruit when thoroughly taught requires a certain amount of experience in field operations or manoeuvres. This he would obtain during the summer immediately following upon the recruit training; for the three months of summer, or of summer and autumn, ought to be devoted almost entirely to field exercises and manoeuvres. If the soldier is then called out for manoeuvres for a fortnight in each of four subsequent years, or for a month in each of two subsequent years, I believe that the lessons he has learned of operations in the field will thereby be refreshed, renewed, and digested, so as to give him sufficient experience and sufficient confidence in himself, in his officers, and in the system to qualify him for war at any moment during the next five or six years. The additional three months' manoeuvre training, beyond the mere recruit training, appears to me indispensable for an army that is to be able to take the field with effect. But that this period should suffice, and that the whole training should be given in nine or ten months of one year, followed by annual periods of manoeuvre, involves the employment of the best methods by a body of officers steeped in the spirit of modern tactics and inspired by a general staff of the first order. The question what is the shortest period that will suffice to produce cohesion belongs to educational psychology. How long does it take to form habits? How many repetitions of a lesson will bring a man into the condition in which he responds automatically to certain calls upon him, as does a swimmer dropped into the water, a reporter in forming his shorthand words, or a cyclist guiding and balancing his machine? In each case two processes are necessary. There is first the series of progressive lessons in which the movements are learned and mastered until the pupil can begin practice. Then follows a period of practice more or less prolonged, without which the lessons learned do not become part of the man's nature; he retains the uncertainty of a beginner. The recruit course of the British army is of four months. A first practice period of six months followed by fresh practice periods of a month each in two subsequent years or by four practice periods of a fortnight each in four successive years are in the proposals here sketched assumed to be sufficient. If they were proved inadequate I believe the right plan of supplementing them would be rather by adding to the number and duration of the manoeuvre practices of the subsequent years than by prolonging the first period of continuous training.