At first, as I have observed, the trunk is commonly omitted. The indifference of the young mind to this is seen in the obstinate persistence of the first scheme of a head set on two legs, even when two arms are added and attached to the sides of the head. Indeed a child will sometimes complete the drawing by adding feet and hands before he troubles to bring in the trunk (see Fig. 8). From this common way of spiking the head on two forked or upright legs there occurs an important deviation. The contour of the head may be left incomplete, and the upper part of the curve be run on into the leg-lines, as in the accompanying example by a Jamaica girl (Fig. 8). [Illustration: Fig. 7 (_a_).] [Illustration: Fig. 7 (_b_).] [Illustration: Fig. 8.] [Illustration: Fig. 9 (_a_).[13]] [Illustration: Fig. 9 (_b_).] [Illustration: Fig. 9 (_c_).] [Illustration: Fig. 9 (_d_).] [13] Fig. 9 (_a_) is a reproduction of a drawing of a girl of four and a half years, from Mr. Lukens' article. The drawing of the trunk may commence in different ways. Sometimes a lame attempt is made to indicate it by leaving space between the head and the legs, that is, by not attaching the legs to the head. Another contrivance is where the space between the legs is shown to be the trunk by shading or by drawing a vertical row of buttons. In other cases the contour of the head appears to be elongated so as to serve for head and trunk. A better expedient is drawing a line across the two vertical lines and so marking off the trunk (see Fig. 9 (_a_) to (_d_)). In drawings made by Brazilian Indians we see another device, _viz._, a pinching in of the vertical lines (see Fig. 9 (_e_)). After the trunk has been recognised by the young draughtsman he is apt to show his want of respect for it by making it absurdly small in proportion to the head, as in Fig. 10. It assumes a variety of shapes, triangular, rectangular, and circular or ovoid, this last being, however, the most common. [Illustration: Fig. 9 (_e_).] [Illustration: Fig. 10.] [Illustration: Fig. 11.] At this stage there is no attempt to show the joining on of the head to the trunk by means of the neck. When this is added it is apt to take the exaggerated look of caricature, as in Fig. 11. A curious feature which not infrequently appears in this first drawing of the trunk is the doubling of the corporeal ovoid, one being laid upon the other. As this appears when a neck is added it looks like a clumsy attempt to indicate the pinch at the waist--presumably the female waist (see Fig. 12). The introduction of the arms is very uncertain. To the child, as also to the savage, the arms seem far less important than the legs, and are omitted in rather more than one case out of two. After all, the divine portion, the head, can be supported very well without their help. [Illustration: Fig. 12.] [Illustration: Fig. 13.]