The cover is then taken away rapidly for one moment, and then as rapidly replaced. The rapidity of this procedure must be such that no movement of the eye, or wandering of the attention over the field of vision, can take place, as long as the diagram remains uncovered.[1] Each time we repeat the experiment a new diagram must be chosen, otherwise the individual momentary impression will supplement the preceding ones. If we wish to obtain unambiguous results we must choose conditions which exclude such influences of previous perceptions. Our question will therefore be limited to this: What is the number of simple and new impressions in consciousness that the focus of attention can grasp in one given moment? In reference to this way of stating the question, an objection to our method of experimenting might be raised. It might be objected that a letter is not a simple element of consciousness, and that we ought rather to use simpler objects, e.g. dots. But since these lack all means of differentiation, the carrying out of the experiment would be rendered much more difficult, if not impossible. On the other hand, we must not forget that our familiarity with letters is of the greatest importance. Because of this a letter of ordinary print can be perceived as quickly as a single dot--a fact any one can easily prove for himself by means of observation. Such symbols, because of their characteristic differences, have this advantage, that after a momentary impression they can be easily retained in consciousness, and thus an account of what has been clearly perceived can be given after the experiment. If we carry out the experiments in the manner described, it appears that an unpractised observer can perceive, at most, only 3-4 letters. After a few more experiments this number increases to 6. Of course, as before mentioned, a new diagram must be used in every new experiment. This value 6 cannot be increased by further practice, and it remains the same for different observers. We are therefore entitled to regard it as a constant for attention for the human consciousness. [Footnote 1: To carry out such experiments more exactly and more uniformly it is best to make use of the simple apparatus called the tachistoscope. A falling screen exposes the object to sight for a very short time, which can be accurately measured. Still, if this apparatus cannot be procured, the procedure described above suffices. Special practice should be devoted to covering and uncovering the diagram, so that this may be done as rapidly as possible.]