Short Report Prospecting for Prehistoric Gardens: Results of a Pilot Study KEVIN C. NOLAN * † Applied Anthropology Laboratories, Department of Anthropology, Burkhardt Buliding Rm 314, Ball State University, Muncie, Indiana 47306 ABSTRACT Non-intensive agricultural systems leave little physical trace on the surface of the landscape. Geochemical analyses are useful in analysing known fields, but often cost-prohibitive for prospection. A method of soil phosphate and magnetic susceptibility survey is proposed as a solution. Either or both soil characteristics should be altered by most agricultural systems. Anomalies in these two soil properties combined with archaeological data indicate at least one possible garden detected in the initial pilot study plot in central Indiana. The detected potential gardens match, in size and distribution, ethnographic accounts of non-intensive agricultural fields. The promise of the proposed method of prospecting for gardens for the study of prehistoric impacts on our inherited landscapes is enormous. Copyright © 2013 John Wiley & Sons, Ltd. Key words: Phosphate; magnetic susceptibility; prehistory; swidden; gardens; eastern North America Introduction Non-intensive agricultural systems leave little imprint on the surface of the modern landscape as their loca- tions are all but invisible to the modern archaeologist. In the absence of significant landscape modifications such as terraces or ridged fields we must rely on other methods of discovery. The effects of agriculture on soils are varied, but well studied (e.g. Eidt, 1977; Sandor et al., 1990; Ball and Kelsay, 1992; Coultas et al., 1993; Kauffman et al., 1993; Miller and Gleason, 1994; Lawrence et al., 1998; Leonardi et al., 1999; Morisada et al., 2000; Ohta et al., 2000; Beach et al., 2002; Sombroek et al., 2002; Branch et al., 2007; Wilson et al., 2008; Alekseeva et al., 2011; Costa et al., 2013). Even slight alterations of soil properties are remark- ably persistent (e.g. Hejcman et al., 2013). Many studies have exploited geochemistry to analyse agricultural activities at known fields, but these techniques are cost-prohibitive for deployment in large-scale survey. A cost-effective method of locating possible non-intensive agricultural plots is desirable. Once suspected gardens are identified, selective deployment of more expensive techniques can confirm or reject the interpretation. If the two types of measurements are highly correlated, the higher-density measurements can be used in geostatistical modelling (i.e. co-kriging) of the more expensive measurements (Burrough and McDonnell, 1998, pp. 147–149). I present the promising results of the first pilot study to locate possible fields where more traditional interpretive geochemistry can be used to assess the nature of these sites. Non-intensive agricultural fields There are few archaeologically identified fields in eastern North America. The general picture for non- intensive gardens is of a variety of plots ranging from about half an acre (e.g. Behrens, 1989; Minnis, 1992) to up to more than 10 acres (Will and Hyde, 1917, p. 106; Wilk, 1997) with most between 1 and 3 * Correspondence to: K. C. Nolan, Applied Archaeology Laborato- ries, Department of Anthropology, Ball State University. E-mail: kcnolan@bsu.edu † The author declares no conflict of interest associated with this work. Copyright © 2013 John Wiley & Sons, Ltd. Received 5 March 2013 Accepted 15 September 2013 Archaeological Prospection Archaeol. Prospect. 21, 147–154 (2014) Published online 4 October 2013 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/arp.1465