Nevertheless, as we have seen, the best evidence attainable points to the conclusion that this simple form of the laughter of social play was preceded by, and grew out of, a less specialised kind of laughter, that of sudden accession of pleasure. We may conjecture that the laughter provoked by tickling was reached in the evolution of our race soon after this reaction passed out of its primal and undifferentiated form as a general sign of pleasurable excitement, and began to be specialised as the expression of mental gaiety and of something like our hilarity. The fact, noted above, {185} that children only laugh in response to tickling when they are in a pleasurable state of mind seems to confirm the hypothesis that the love of fun, which is at the bottom of tickling and makes it perhaps the earliest clear instance of mirthful play with its element of make-believe, first emerged gradually out of a more general feeling of gladness. {186} CHAPTER VII. DEVELOPMENT OF LAUGHTER DURING THE FIRST THREE YEARS OF LIFE. Having examined the earliest and distinctly hereditary germs of the laughing impulse in the child, we may pass to the consideration of its expansion and specialisation during the first years. Although, so far as I am aware, the new child-study has not yet produced a methodical record of the changes which this interesting expression of feeling undergoes, we may by help of such data as are accessible be able to trace out some of the main directions of its development. Two closely connected problems are involved here: (_a_) how the expressive movements, the laugh and the smile, themselves change and get differentiated; and (_b_) how the psychical process which precedes and excites these expressive movements grows in complexity and differences itself into the various forms of gaiety or amusement enumerated above. In dealing with these early manifestations we shall, of course, look for reactions which are spontaneous, in the sense of not being due to imitation and the lead of others. Yet it will not always be easy to determine what are such. It has been pointed out above that laughter is one of the most contagious of the expressive movements. Children, therefore, who are much given to imitation may be {187} expected to show this contagiousness in a particularly clear manner.