If this is a correct analysis of the experience of the tickling which excites laughter, we seem to have in it at a very early age elements which are to be found, in a more fully developed form, in the later and more complex sorts of mirth, namely, relief from a serious and constrained attitude, a transition from a momentary apprehension induced by the presentation of the partially unknown, to a joyous sense of harmless make-believe. That this is so is further evidenced by the familiar fact that a child, when used to the game, will begin to laugh vigorously when you only threaten with the advancing fingers. As a German writer observes, this is a clear case of Lipps’ theory of annihilated expectation;[43] only he omits to note that the laughter depends, not on the mere fact of annihilation, but on the peculiar conditions of it in this case, involving a slight shock at the approach of something partially unknown to a specially sensitive region of the organism, and the instant correction of the apprehension by a recognition of its harmlessness. Much the same kind of stimulative process seems to be present in the other and allied cases of reflex or quasi-reflex laughter. It is well known that certain sense-stimuli which excite sensations of a disagreeable character, but which, though acute, are not violent, such as the application of a cold douche, are apt to provoke laughter. According to the German authority just quoted, the effect depends here, too, on variation in respect of the intensity and the locality of the stimulation. He found further, in carrying out psychological experiments, that whereas the introduction of a stronger stimulus than was expected is apt to excite apprehension in {65} the subject, that of a weaker stimulus will excite laughter.[44] Here, too, we seem to have a sensational reflex in which is present a distinctly mental element, _viz._, a moment of mild shock and apprehension at the sudden coming of something disagreeable and partially unknown, instantly followed by another moment of dissolution of shock in a pleasurable recognition of the harmlessness of the assault. 2. Laughter is not, however, always of this reflex form. It may arise without sensory stimulation in an “automatic” manner as the result of a cerebral rather than of a peripheral process. This is illustrated by the seemingly causeless laughter which breaks out in certain abnormal states and has an “uncanny” aspect for the sane observer. A well-known example of this is the effect of the action on the brain centres of laughing gas and other substances. Such “automatisms” occur, however, within the limits of normal experience, as when a person laughs during a state of high emotional tension. I propose to speak of such seemingly uncaused reactions as _nervous laughter_.[45]