As far as possible I have sought spontaneous drawings of quite young children, _viz._, from between two and three to about six. In a strict sense, of course, no child's drawing is absolutely spontaneous and independent of external stimulus and guidance. The first attempts to manage the pencil are commonly aided by the mother or other instructor, who, moreover, is wont to present a model drawing, and, what is even more important at this early stage, to supply model-movements of the arm and hand. In most cases, too, there is some slight amount of critical inspection, as when she asks, "Where is papa's nose?" "Where is doggie's tail?" In one case, however, I have succeeded in getting drawings of a little girl who was carefully left to develop her own ideas. Even in the instances where adult supervision is apt to interfere, we can, I think, by patient investigation distinguish traits which are genuinely childish. A child's drawing begins with a free aimless swinging of the pencil to and fro, which movements produce a chaos of slightly curved lines. These movements are purely spontaneous, or, if imitative, are so only in the sense that they follow roughly the directions of another's pencil. In this first line-scribble there is no serious intention to trace a particular form. What a child seems to do in this rough imitation of another's movements is to make a tangle of lines, more or less straight, varied by loops, which in a true spirit of play he makes believe to be the semblance of "mamma," "pussy," or what not, as in Fig. 1 (_a_) and (_b_). Possibly in not a few cases the interpretation first suggests itself after the scribble, the child's fancy discerning some faint resemblance in his formless tangle to a human head, a cat's tail, and so forth. [Illustration: Fig. 1 (_a_).[11]] [Illustration: Fig. 1 (_b_).[11]] [11] Fig. 1 (_a_) is a drawing of a man by a child of twenty months, reproduced from Prof. M. Baldwin's _Mental Development_, p. 84; Fig. 1 (_b_) is a drawing of a man by a child of two years three months, reproduced from an article on children's drawings by Mr. H. T. Lukens in _The Pedagogical Seminary_, vol. iv. (1896). This habit of scribble may persist after a child attempts a linear description of the parts of an object. Thus a little girl in her fourth year when asked to draw a cat produced the two accompanying figures (Fig. 2 (_a_) and (_b_)). [Illustration: Fig. 2 (_a_).] [Illustration: Fig. 2 (_b_).] Here it is evident we have a phase of childish drawing which is closely analogous to the symbolism of language. The form of representation is chosen arbitrarily and not because of its likeness to what is represented. This element of symbolic indication will be found to run through the whole of childish drawing.