Seth Mallios Back to the Bowl: Using English Tobacco Pipebowls to Calculate Mean Site-Occupation Dates ABSTRACT The form of English white ball-clay pipebowls can be used to determine accurate mean dates for historical contexts. Using previously established bowl seriations and typologies on the well-dated material assemblage uncovered at the site of the 1607 James Fort at Jamestown Island, Virginia, a newly developed calculation proves to be a reliable technique of assessing chronology. Determining the pipebowl mean date involves identifying the shape of each bowl, counting the number of examples of each morphological type, and then completing a series of simple arithmetic calculations. The pipebowl-dating device correlates well with other archaeo- logical lines of evidence. On average, pipebowl mean dates are within seven years of mean dates established by other factors. This method regularly outperforms established pipestem-based mean-date measures. Introduction English white ball-clay pipebowls can be used to determine reliable mean dates for historical features. Analyses of recent material discoveries at the site of the original 1607 James Fort indicate that English pipebowls offer mean dates for 17th-century archaeological contexts that are corroborated by other factors. Continuing excavations by the Jamestown Rediscovery team on the Association for the Preservation of Virginia Antiquities' (APVA) Jamestown Island property from 1994 to 2000 produced multiple narrowly dated features that are used to test previously established and newly created pipe-dating methods. These include temporal measures developed in the past by Lewis Binford (1962), Lee Hanson (1969), and Robert Heighton and Kathleen Deagan (1971). Based on bowl typologies established using seriation by David Atkinson and Adrian Oswald (1969) and Oswald (1951, 1975), the calculation for pipebowl mean dates presented here proves to be a reliable technique for determining chronology. On average, it provides occupation midpoints within seven years of mean dates established by other archaeological and historical lines of evidence. The bowl-based method consistently outperforms commonly used stem- based mean-date measures. When used on material uncovered at 44JC802, an off-island 17th-century site in Jamestown's historical hinterland, the pipebowl mean date confirms the likely identity of the site's landowner during its occupation. These preliminary results encourage additional use and testing of the dating technique at other historical sites. Beginning with a review of pipe-based chro- nologies developed over the last 50 years, which, with few exceptions (Oswald 1951), focused more on pipestems than bowls (Har- rington 1954; Binford 1962; Hanson 1969; Heighton and Deagan 1971; Deetz 1987), the following analysis also offers a brief overview of the history and archaeology of the first per- manent English settlement in America as back- ground. The four steps involved in determining a pipebowl mean date are then detailed. This temporal tool produces more accurate results than various stem-based measures on the James- town assemblage. Subsequent use of the dating device on neighboring sites further demonstrates its relative accuracy and utility. Throughout the article, the objects of study are English white ball-clay pipes-often called kaolin-and do not include terra-cotta "Chesapeake" or "Colono" pipes (Walker 1972:161). On the basis of mak- er's marks and motifs present on the bowls and stems, the overwhelming majority of the white ball-clay pipes considered in this analysis are identified as English and not Dutch. Ultimately, even with small assemblages containing only 6 to 15 bowls, this chronological method proves to be a reliable material line of evidence with which to examine archaeological patterns. Background Pipe-Dating Methods Before J. C. Harrington's landmark discovery regarding the correspondence between time and Historical Archaeology, 2005, 39(2):89-104. Permission to reprint required. Accepted for publication 16 March 2004.