Although periods of incubation are essential for every specific habit, practically every act of skill is dependent upon a number of simpler habits. At any one time progress may be made in utilizing some of these habits, even though others could not be advantageously hastened. Thus the period of incubation should not necessarily cause any profound slump in the advance. Almost invariably, however, it produces a plateau which persists until the worker <p 248> has mastered the expert way. The golf player, for example, usually finds he is able to drive longer and straighter balls at the beginning of the season than a little later. The reason is that in golf the perfect stroke is the product of almost automatic muscular action. In the first round the swing of the driver or iron is not consciously governed, and the muscular habit of the previous year controls. Later, as the player concentrates on his task of correcting little faults or learning more effective methods, his stroke loses its automatic quality, his game falls off, and it is not until he masters his new form that he attains high efficiency. The same cycle is repeated in office and factory operations, where efficiency is possible only when the hands carry out automatically the desired action. In typewriting and telegraphy, in the handling of adding machines, in the feeding of drill presses, punch presses, and hundreds of special machines, the learner passes through three distinct phases: first, swift improvement in which prelearned move- <p 249> ments and skill are brought to bear on the task under the stimulus of both novelty interest and voluntary interest; second, arrested progress-- the period of incubation or habit formation; and the final stage of automatic skill and efficiency. _Since increase of efficiency is dependent upon continued efforts of will, slumps are inevitable. Voluntary attention cannot be sustained for a long period_. Work requiring effort is always subject to fluctuations. The man with a strong will may make the lapses in attention relatively short. He may be on his guard and ``try to try'' most faithfully, but no exertion of the will can keep up a steady expenditure of effort in any single activity. All significant _*increases_ in efficiency, however, are dependent upon voluntary attention--upon extreme exertions of the will.