the thought-process or the composition of the work of art may undergo deviations or additions in its separate parts. The regularity of the process as a whole remains undisturbed by this. A work of art is just as little a mere product of association as is a thought arranged in sentences. Various phenomena of everyday experience find their explanation in these psychological observations. First of all must be mentioned the seldom-noted fact, that we are able in our speech to bring to an end a fairly complicated thought without difficulty, although at the beginning of the sentences we are not at all clear as to the separate words and ideas or their combinations. Some people, when they are obliged to speak in public, fail simply because their confidence in this self-regulation of the train of thoughts is lacking at such moments. And this again is due to the fact that they think they must first of all find the suitable transition from one word to the next. In free conversation they can carry to an end without a break the most complicated sentences, while in public their speech is hesitating and embarrassed, and they are every moment in danger of breaking down. In such a case absolute confidence in the possibility of expressing freely and involuntarily the thought in one's mind is the surest help to overcome these difficulties. Of course a sensible training will also help. Let us call to mind the processes by means of which the beginning and end of an expression of thought are held together into one sentence or into several sentences joined together by the same thought. We note at once that the general content in its whole feeling-quality is already present as soon as the first word is spoken, while the ideas and the corresponding words are not clearly in consciousness beyond that first beginning. If the process continues without associative distractions and additions, by which, occasionally, parts that lie far from the original thought are added to it, then we notice at the same time that that beginning feeling corresponds perfectly with the terminal feeling that accompanies the termination of the spoken thought. This terminal feeling is generally at first much stronger than the initial feeling, but then it gradually goes over into the feeling-quality that is preparing the next thought. Now it is obvious at once that all these phenomena correspond in essentials with those we observed in our metronome experiments. In these experiments the conditions were much more simple and exact, so that they strengthen the more uncertain observations in ordinary reading and thinking.