The effect of this ingredient of the glaze whatever its composition may have been is easily understood. All porcelain and pottery undergoes a considerable amount of contraction--from loss of moisture, etc.--in the kiln, and to obtain a perfectly even glaze it is necessary that the contraction of the glaze should be the same as that of the body. Clearly this ingredient caused the glaze to contract to a greater extent than the body, and so to split up into minute fissures. The Chinese were able to control to a great extent the size and nature of the crackle, as is shown by the appearance of alternate bands of large and small crackle on the same piece. The methods of colouring the crackle include rubbing red ochre, ink, and decoction of tea leaves into the cracks before the ware was quite cool. Another method is described by Bushell (_O. C. A._, p. 511) by which a white crackled ware was stained pink or crimson. The vessel was held in the fire in an iron cage until thoroughly heated, and then water mixed with gold-pink colouring matter was blown on to it. This, however, is a later process. Most of the monochrome glazes are occasionally crackled, but the most characteristic colours of the crackle glazes are the greyish white (the _blanc un peu cendré_ of Père d’Entrecolles), and light buff, which were probably intended to recall the ash colour (_hui sê_) and the millet colour (_mi sê_) of the Sung _Ko yao_. Some of the light buff or “oatmeal” crackles of the early Ch’ing period are peculiarly refined and beautiful. Though this has seemed a favourable opportunity for discussing crackle glazes it is not to be supposed that they were a speciality of the K’ang Hsi period. They are common to every age since the Sung dynasty, and probably they were never made in such abundance and with such care as in the Yung Chêng and early Ch’ien Lung periods. CHAPTER XII YUNG CHÊNG [chch 2] PERIOD (1723–1735)