the vessel. In a similar manner, clay cores, wax layers, and ornaments were fabricated before the construction of the wax runners and vents. Thereafter, the multi-layered mold was applied, dried, cut with “wedge-shaped mortises” along the edges, and lifted from the model in pieces. After the last-made pieces had thoroughly dried, the mold was opened and every part of the wax was physically taken away (not melted away). After the final inspection and any necessary repairs, the mold assembly was reconstructed with an additional coarse, thick layer, dried gradually, and baked hard before the form was filled with molten metal (“Fine Metal-work,” 628–630). In addition, Maryon and Plenderleith considered their reconstruction to be a variation of the lost-wax process: “That these moulds were piece moulds demonstrates also that the founder wished to lift them from the model and to remove the wax by hand. Had he wished to employ the cire perdue process there would have been no need for the piece-mould—the mould could have been made as a whole” (“Fine Metal-work,” 630). This excessively complex method, whether a variation of the lost-wax process or not, was impractical and unrealistic. The proposal of such an improbable method was likely due to the authors’ deeply entrenched a priori belief in the use of wax models. At the same time, they must have known the already published Anyang materials that included the mold fragments: containing mortises and tenons, these fragments were hardly likely to be pieces broken from a wholly formed investment mold. They must have instead come from molds constructed in fitted sections.

34. Li, “Fushen Zang,” 476; Su, “Ershi Shiji dui Xianqin Qingtong Liqi Zhuzao Jishu de Yanjiu,” 388.

35. For critical analysis of this issue, see Bagley, Shang Ritual Bronzes in the Arthur M. Sackler Collections, 27–28.

36. Karlbeck, “Anyang Moulds,” 42–43.

37. Preferring the section-mold theory, Karlbeck, being cautious and circumspect, still did not rule out the possibility of the direct lost-wax process: “There is also the possibility that they [the section molds] were used in the cire perdue method, the sections being formed round a wax model which was melted out and replaced by molten bronze” (“Anyang Moulds,” 43).

38. Karlbeck, “Anyang Moulds,” 44.

39. Ibid., 46.

40. See Guo, Ruixian Yiqi, 19; Rong, Shangzhou Yiqi Tongkao, 158.

41. Chen, “Yindai Tongqi Sanpian,” 36–43.