Page:Brinkley - Japan - Volume 7.djvu/306

 was practised by them with much success. But they treated the guard as though it were a block of cardboard, and were content with the simple operation of piercing, so that the decorative design appeared in outline only. At the end of the sixteenth century, or the beginning of the seventeenth, a new departure was made by adding surface modelling to pierced work. The difference thus produced can be easily explained by saying that whereas a design of cherry petals, for example, took the form of a mere diaper according to the old method, it became, according to the new, a cluster of accurately shaped blossoms and leaves suspended within the circumference of the guard. Under this artistic impulse the guard soon ceased to have the character of a frame, or field, for the design, and was wholly absorbed into the latter. An immense variety of beautiful and cleverly conceived specimens then came into existence. The rim of the guard, ceasing to be rigidly circular, square, or oval, adapted itself to the demands of the design; and the carver, while taking care not to sacrifice the protective purpose of his work, allowed himself wide latitude and irregularity of shape. Thus the "ascending" and "descending" dragons, together with the clouds among which they fly, were disposed so that the backs of the monsters formed the rim of the guard; and a procession of rats pursuing each other in a circle filled all the space surrounding a central haft-socket; or a branch of cherry-bloom, or of plum-blossoms, or of pine-branches, or a cluster of all three combined, was skilfully bent into a circular medallion. Wreaths of iris, sheaves of rice, circlets of intertwined serpents, loops of crayfish, garlands of bean-sprays,—it would scarcely be possible to enumerate the