Both contain the same number of rises and falls, but in a different arrangement. A shows a pronounced example of a descending row of beats, B a similar example of a row that first ascends and then descends. With a suitable rapidity of the metronome we can easily hear at will into the uniform beats of the pendulum each of these rhythms. If, however, we have once made our choice between the two forms, then we group the beats that follow the row A in exactly the same manner as the row A, and the same thing happens with the row B. Such a spontaneous repetition is only possible owing to the fact that at the last beat of each row we group the whole together. This we do with the succeeding beats as well, just as we have seen to be generally the case in measuring the scope of consciousness. Now if we observe our feelings we obtain an important addition to our previous observations. They showed us that a very important part of such a process was composed of the alternating feelings of strain and relaxation, and perhaps also of excitation and quiescence, and lastly of agreeableness. This last feeling was especially strong at the end of a row of beats, caused by the arrangement of the single element into one rhythmically ordered whole. It is obvious now that the centre of gravity of the affective process lies every time at the end of a row, where the superimposed rhythmical feelings run together into one unity. For it is unmistakably this feeling that allows us directly to apprehend the succeeding rows as identical with the preceding ones in a succession of similar rows. What we apperceive is not the preceding row itself. The greater number of its elements lie already in the darker field of consciousness. We apperceive rather this aggregate feeling, which is joined to the last directly apperceived element, and which is the resultant of the preceding affective processes. Now let us compare this terminal feeling, that lends a given rhythm its essential and peculiar affective character, as it appears in the two examples represented by A and B. It is evident that however much on the one hand a row may depend upon the constitution and the arrangement of the preceding components, it yet on the other hand always possesses its own specific quality. It is true that we can always classify this under one or more of the six chief qualities, and yet we do not thereby account for its own peculiar quality, which differentiates it from the others of the same class. It also cannot be considered a mere summation of the simple feelings that axe joined to the separate parts of the process. The feelings of strain and relaxation that are distributed over the rows A and B are the same. They differ at most in the degree of intensity. We cannot therefore understand why the feelings that remain behind at the end of each row should be so different. But it is so. We can convince ourselves of this more directly than in the experiments with voluntary rhythmical