Development and evaluation of geochemical methods for the sourcing of archaeological maize

Authors

L.V. Benson, U.S. Geological SurveyFollow
H. E. Taylor, U.S. Geological SurveyFollow
K.A. Peterson, Chaco Culture National Historical Park
B.D. Shattuck, Chaco Culture National Historical Park
C.A. Ramotnik, U.S. Geological Survey
J.R. Stein, The Navajo Nation Chaco Sites Protection Program

Document Type

Article

Date of this Version

2008

Citation

Journal of Archaeological Science 35 (2008) 912-921; doi:10.1016/j.jas.2007.06.018

Abstract

Strontium (Sr)-isotope values on bone from deer mice pairs from 12 field sites in the Chaco Canyon area, New Mexico, were compared with isotope values of synthetic soil waters from the same fields. The data indicate that mice obtain Sr from near-surface sources and that soil samples collected at depths ranging from 25 to 95 cm contain Sr that is more accessible to the deep roots of maize; thus, synthetic soil solutions provide better data for the sourcing of archaeological maize. However, the Sr-isotope composition of mice may be more valuable in sourcing archaeological remains of animals such as rabbit, turkey, and deer.

In a separate study, five Native American maize (Zea mays L. ssp. mays) accessions grown out at New Mexico State University Agricultural Science Center, Farmington, New Mexico were used to determine if soil-water metal pairs partition systematically into cobs and kernels. The sampled maize included landraces from three Native American groups (Acoma, Hopi, Zuni) that still occupy the Four Corners area. Two cobs each were picked from 10 plants of each landrace. Partitioning of the Ba/Mn, Ba/Sr, Ca/Sr, and K/Rb metal pairs from the soil water to the cob appears to behave in a systematic fashion. In addition, 51 rare earth element (REE) pairs also appear to systematically partition from the soil water into cobs; however, the ratios of the REE dissolved in the soil waters are relatively invariant; therefore, the distribution coefficients that describe the partitioning of REE from the soil water to the cob may not apply to archeological cobs grown under chemically heterogeneous conditions. Partitioning of Ba/Rb, Ba/Sr, Mg/P, and Mn/P metal pairs from the soil water to kernels also behaves in a systematic fashion. Given that modern Native American landraces were grown under optimal environmental conditions that may not have been duplicated by prehistoric Native Americans, the distribution coefficients obtained in this study should be used with caution.