The theory has been advanced that geometric ornament developed through the degeneration of perspective designs; in part perhaps also through that of symbolic designs. It is assumed that the symbol, or the object represented was misunderstood and that in course of time through a process of slurring, by careless and inaccurate representation the forms became fragmentary and finally lost all semblance to the original. It is not possible to accept this theory, because the conditions under which the supposed slurring occurs are seldom realized. Slovenly work does not occur in an untouched primitive culture. Misunderstandings may happen in cases of borrowing of designs or in that of a gradual transformation of those concepts that find expression in decorative art. Actual slurring is found in factory production. By an examination of a few cases of this kind we were able to show that it does not lead to geometrization, but to the growth of an individualism akin to that of our handwriting. It cannot be denied that in such cases occasion for re-interpretation with consequent changes of form occur, but these are not frequent. On the other hand we were able to show that reading in of realistic meanings into geometric forms is quite common. We proved this by means of a detailed comparison of the style of painting and embroidery of the North American Plains Indians which we found to be practically identical everywhere, while the interpretations varied from tribe to tribe. This phenomenon agrees with the general tendency to keep intact the form, but to endow it with new meaning according to the chief cultural interests of the people. We pointed out the prevalence of the same tendency in folk-tales and ritual. As a general explanation the geometrization of realistic patterns is, therefore, unacceptable. In the majority of cases it seems to be rather due to the inclination of man to give a meaning to geometric form, as we enjoy reading meanings into the forms of clouds and mountains. We were also able to describe a few cases in which the process of reading in has actually been observed. Another fact prevents us from considering geometrization as a general historical process. It is very seldom only that the steps are found so distributed that they can be proved to follow one another in time. Much oftener all are found at the same time among the same people. Considering all these points we reached the conclusion that the stylistic form which contains to a greater or lesser extent constant geometrical elements, is decisive in determining the manner in which representations are rendered. We were thus led to the attempt to find the principles underlying art styles.