It may be thought by some that the processes of development indicated above are insufficient and unsatisfactory. There are those who, seeing these forms already endowed with symbolism, begin at what I conceive to be the wrong end of the process. They derive the form of symbol directly from the thing symbolized. Thus the current scroll is, with many races, found to be a symbol of water, and its origin is attributed to a literal rendition of the sweep and curl of the waves. It is more probable that the scroll became the symbol of the sea long after its development through agencies similar to those described above, and that the association resulted from the observation of incidental resemblances. This same figure, in use by the Indians of the interior of the continent, is regarded as symbolic of the whirlwind, and it is probable that any symbol-using people will find in the features and phenomena of their environment, whatever it may be, sufficient resemblance to any of their decorative devices to lead to a symbolic association. [Illustration: FIG. 481.--Theoretical development of the current scroll.] One secret of modification is found in the use of a radical in more than one art, owing to differences in constructional characters. For example, the tendency of nearly all woven fabrics is to encourage, even to compel, the use of straight lines in the decorative designs applied. Thus the attempt to employ curved lines would lead to stepped or broken lines. The curvilinear scroll coming from some other art would be forced by the constructional character of the fabric into square forms, and the rectilinear meander or fret would result, as shown in. Fig. 482, _a_ being the plain form, painted, engraved, or in relief, and _b_ the same idea developed in a woven fabric. Stone or brick-work would lead to like results, Fig. 483; but the modification could as readily move in the other direction. If an ornament originating in the constructional character of a woven fabric, or remodeled by it, and hence rectilinear, should be desired for a smooth structureless or featureless surface, the difficulties of drawing the angular forms would lead to the delineation of curved forms, and we would have exactly the reverse of the order shown in Figs. 482 and 483. The two forms given in Fig. 484 actually occur in one and the same design painted upon an ancient Pueblo vase. The curved form is apparently the result of careless or hurried work, the original angular form, having come from, a textile source. [Illustration: _a_, free-hand form. _b_, form imposed, by fabric. FIG. 482.--Forms of the same motive expressed in different arts.] [Illustration: _a_, free-hand form. _b_, form imposed by masonry. FIG. 483.--Forms of the same motive expressed in different arts.] [Illustration: _a_ _b_ FIG. 484.--Variations resulting from change of method.]