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 at a considerable distance the movements of the mother’s pencil.[230] They may be made expressive or significant in two ways. In the first place, a child may by varying the swinging movements accidentally produce an effect which suggests an idea through a remote resemblance. A little boy when two years and two months, was one day playing in this wise with the pencil, and happening to make a sort of curling line, shouted with excited glee, ‘Puff, puff!’ _i.e._, smoke. He then drew more curls with a rudimentary intention to show what he meant. In like manner when a child happens to bend his line into something like a closed circle or ellipse he will catch the faint resemblance to the rounded human head and exclaim, ‘Mama!’ or ‘Dada!’ ----- Footnote 230: E. Cooke gives illustrations of these in his thoughtful and interesting articles on “Art-teaching and Child-nature,” published in the _Journal of Education_, Dec., 1885, and Jan., 1886. ----- But intentional drawing or designing does not always arise in this way. A child may set himself to draw, and make believe that he is drawing something when he is scribbling. This is largely an imitative play-action following the direction of the movements of another’s hand. Preyer speaks of a little boy who in his second year was asked when scribbling with a pencil what he was doing and answered ‘writing houses’. He was apparently making believe that his jumble of lines represented houses.[231] Almost any scribble may in this earliest stage take on a meaning through the play of a vigorous childish imagination. ----- Footnote 231: Preyer, _op. cit._, p. 47. ----- [Illustration: Fig. 1 (_a_) and (_b_).] The same play of imagination is noticeable in the child’s first endeavours to draw an object from memory when he is asked to do so. Thus a little girl in her fourth year referred to by Mr. E. Cooke when asked to draw a cat produced a longish irregularly curved line crossed by a number of shorter lines, which strange production she proceeded quite complacently to dignify by the name ‘cat,’ naming the whiskers, legs, and tail (Fig. 1 (_a_); compare the slightly fuller design in Fig. 1 (_b_)).[232] ----- Footnote 232: Taken from E. Cooke’s articles already quoted, drawings 19 and 20. ----- Here it is evident we have a phase of childish drawing which is closely analogous to the symbolism of language. The representation is arbitrarily chosen as a symbol and not as a likeness. This element of a non-imitative or symbolic mode of representation will be found to run through the whole of childish drawing.