The disregard of the original pattern is such that in some specimens (fig. 15) part of the design has been cut off in order to make the sides fit together. In our illustration the parts cut off,—the right upper corners,—have been reconstructed. It might be said that similar conditions prevail in modern, patterned fabrics that are made into garments. In this case the manufacturer tries to attain a pleasing effect for the fabric as a whole. If economy of material did not interfere, the tailor would fit the pattern together, but we always feel the conflict between the pattern and the requirements of the finished garment. [Illustration: Fig. 16. Fringe from legging, Thompson Indians.] As another example I mention a legging made by an Indian woman from the interior of British Columbia. It bears the usual decoration,—a long fringe along the outer seam. The fringe consists of a long piece of curried skin cut in narrow strips. These strips are decorated in rhythmic order (fig. 16), a string decorated by one glass bead and two bone beads in alternating order is followed by a plain string, next by one decorated with single alternating glass and bone beads, then a plain one and finally one like the first. When we indicate the plain and decorated strips by letters, we find the arrangement . . . | A B C B A | A B C B A | . . . repeated over and over again.[21] The important point to be noted is, that when in use, the fringe hangs down without order along the outer side of the leg so that the elaborate rhythmic pattern cannot be seen. The only way in which the maker can get any satisfaction from her work is while making it or when exhibiting it to her friends. When it is in use there is no esthetic effect. [21] James Teit, The Thompson Indians of British Columbia, Publications of the Jesup North Pacific Expedition, New York 1900, Vol. I, p. 382; see also Gladys A. Reichard, The Complexity of Rhythm in Decorative Art, American Anthropologist N. S., Vol. 24 (1922), p. 198. Other cases occur in which motives are applied that are practically invisible. Thus in mattings from Vancouver Island, the weaver will alternate the direction of the strands in squares without any attempt to set off the surface in colors (fig. 17). When the matting is new these patterns may be seen in reflected light, but after a very short time they disappear almost completely. [Illustration: Fig. 17. Twilled weaving showing alternation of patterns.] Similarly the woven patterns on arm rings from the Ucayali are practically invisible.[22] [22] Max Schmidt, Besondere Geflechtsart der Indianer im Ucayaligebiete, Archiv für Anthropologie, N. S., Vol. VI (1907), p. 270.