turned to the subject. "He stood at the window at B, looking out at a sea mist thoughtfully, and said suddenly: 'Mamma, do you remember I told you that I had seen angels? Well, I want now to say they were not angels, though I thought they were. I have seen it often lately, I see it now: it is bright stars, small bright stars moving by. I see it in the mist before that tree. I see it oftenest in the misty days. . . . Perhaps by and by I shall think it is something in my own eyes." Here we see a long and painstaking attempt of a child's brain to read a meaning into the "flying spots" which many of us know, though we hardly give them a moment's attention.
What are children's first thoughts about their dreams like? I have not been able to collect much evidence on this head. What seems certain is that to the naïve intelligence of the child these counterfeits of ordinary sense-presentations are real external things. The crudest manifestation of this thought-tendency is seen in taking the dream apparition to be actually present in the bedroom. A boy in an elementary school in London, aged five years, said one day, "Teacher, I saw an old woman one night against my bed." Another child, a little girl, in the same school, told her mother that she had seen a funeral last night, and on being asked "Where?" answered quaintly, "I saw it in my pillow." A little boy, whom I know, once asked his mother not to put him to bed in a certain room "because there were so many dreams in the room." In thus materializing the dream and localizing it in the actual surroundings, the child but reflects the early thought of the race which starts from the supposition that the man or animal which appears in a dream actually approaches the sleeper.
The Nature-man, as we know from Prof. Tylor's researches, goes on to explain dreams by his theory of souls or "doubles" ("animism"). Children do not often find their way to so subtle a line of thought. Much more commonly they pass from the first stage of naive acceptance of objects as present here and now to the identification of dreamland with fairyland or the other and invisible world. There is little doubt that the imaginative child firmly believes in the existence of this invisible world, keeps it carefully apart from this one, even though at times he may give it a definite locality in this e. g., in C 's case, in the wall of his bedroom. He gets access to it by shutting out the real world, as when he closes his eyes tightly and "thinks." With such a child dreams get taken up into the invisible world. Going to sleep is now recognized as the surest way of passing into this region. The varying color of his dreams, now bright and dazzling in their beauty, now black and terrifying, is explained by a reference to the division of that fairy world into princes and good
