All =practice= begins in the state of _attention_; but practice, once started, may go on when attention is distracted from the matter in hand. We give a great deal of attention to our first finger-exercises on the piano; presently, if we have continued them long enough, we may practise Chopin on the clavier while we are reading a book or thinking out a problem; the fingers do the practising for themselves. If we follow _the course of practice_, from day to day, we find that improvement is not steady; we gain very quickly at first, then come to a point at which we remain stationary for a while, then make another and slower gain, then rest at a second plateau or level of practice, and so on. It is doubtful, however, whether this stepwise advance is characteristic of practice itself, that is, of the nervous change produced by repeated stimulation of the same nerve-elements; it seems rather to be due to changes in our method of working, to the sudden discovery of some new trick of procedure, or the sudden release from some hampering peculiarity of method. We cannot speak in positive terms since, unfortunately for psychology, the investigators of practice have been more concerned with outward results and practical value than with description of the correlated mental processes. In psychological experiments, _the practised observer has a threefold superiority over the unpractised_: his attitude to the stimuli, in successive observations, is more nearly uniform; his attention is sustained at a higher level; and his discrimination is more refined. This means that the focal mental processes are few in number; that they are extremely vivid; and that they are protected, by strong inhibitory forces, against intrusion from the outside. It is clear therefore that practice is very desirable; but it is clear also that experimental results may be compared only if the stage of practice at which they are obtained is the same. This rule has some odd examples: an observer, for instance, who is practised in the discrimination of lifted weights grows physically stronger with his practice, and may therefore judge quite differently from the unpractised observer.