The thing which strikes one most perhaps in these early fears is how little they have to do with any remembered experience of evil. The child is inexperienced, and if humanely treated knows little of the acuter forms of human suffering. It would seem at least as if he feared not because experience had made him apprehensive of evil, but because he was constitutionally and instinctively nervous, and possessed with a feeling of insecurity. This feeling of weakness and insecurity comes to the surface in presence of what is unknown in so far as this can be brought by the child’s mind into a relation to his welfare—as disturbing noises, and the movements of things, especially when they take on the form of approaches. The same thing is, as we have seen, illustrated in the fear of the dark. A like explanation seems to offer itself for other common forms of fear, especially those excited by others’ threats, as the dread of the policeman, and little George Sand’s horror at the idea of being shut up all night in the ‘crystal prison’ of a lamp. The fact that children’s fears are not the direct product of experience is expressed otherwise by saying that they are the offspring of the imagination. A child is apt to be afraid because he fancies things, and it will probably be demonstrated by statistical evidence that the most imaginative children (other things being equal) are the most subject to fear. In certain of these characteristics, at least, children’s fears resemble those of animals. In both alike fear is much more an instinctive recoil from the unknown than an apprehension of known evil. The shying of a horse, the apparent fear of dogs at certain noises, probably too the fear of animals at the sight and sound of fire—so graphically described by Mr. Kipling in the case of the jungle beasts—illustrate this. Animals too seem to have a sense of the uncanny, when something apparently uncaused happens, as when Romanes excited fear in a dog by attaching a fine thread to a bone, and by surreptitiously drawing it from the animal, giving to the bone the look of self-movement. The same dog was frightened by soap-bubbles. According to Romanes, dogs are frightened by portraits. It is to be added, however, that in certain of animal fears the influence of heredity is clearly recognisable, whereas in children’s fears I have regarded it as doubtful. The fact that a child is not frightened at fire, which terrifies many animals, seems to illustrate this difference.[155] ----- Footnote 155: See Perez, _L’Education dès le berceau_, pp. 96-99. On animal fears, see further Romanes, _Animal Intelligence_, p. 455 f.; Preyer, _op. cit._, p. 127 ff. and p. 135; Perez, _First Three Years of Childhood_, p. 64 ff. -----