No fewer than 648 stem fragments were recovered from the cellar and their stem-hole diameters, using J. C. Harrington's chart,[26] indicated a manufacture date in the period 1680-1710. Because pipes are considered to have had a short life, it is generally assumed that the dates of manufacture and deposition are not far apart. Other artifacts from the deposit, notably the large glass stem, the wine bottles, small wineglass and, of course, the pipe bowl shapes, together suggested a terminal date for the group within the period 1690-1700. Using the Binford formula,[27] the 648 stem fragments suggested a mean date of 1698. Experience has shown that the formula is likely to be accurate to three or four years either way on a sampling of that size.[28] The presence of the same maker's initials, I·F, on pipe bowls at different levels of the cellar fill strongly pointed to a homogeneity of deposition. Although it is impossible to identify the owners of the initials with any certainty, it is worth noting that there was a Josiah Fox making pipes in Newcastle-under-Lyme in and after 1683 whose initials are the same as those most common in the Clay Bank cellar. The I·F mark was somewhat unusual in that it was impressed between two X's across the top of the stem (fig. 16, no. 11). All other marks, save one, were in the normal position, to left and right of the heels. These comprised W F (William Ferry, Marlborough, about 1700?), or perhaps W.P., II I (Henry Jones, London, 1688?)[29] and V R. The remaining mark, S A (fig. 16, no. 14) occurred on the bases of two bowls with neither heels nor spurs. From the oystershell layer south of the existing house came a bowl fragment ornamented with the name of a well-known Bristol pipe-making family, I TIPPET, in a raised cartouche on the side. This was probably Jacob Tippett whose name appeared in the Bristol Freedom Rolls in 1680.[30] In addition to the few marked bowls, two stems were of interest in that they had been ground or pared down to enable the pipes to be used again, one being only 2-1/4 inches in length (fig. 16, nos. 12 and 13). Such frugality might be construed as being associated with a household of small means. Also present were a few brown stem fragments and part of one decorated bowl (fig. 8, no. 9) of Virginia, possibly Indian, manufacture. Conclusions