Differences in isotopic compositions of individual grains and aggregated seed samples affect interpretation of ancient plant cultivation practices

Abstract

The stable carbon ({delta}13C) and nitrogen ({delta}15N) isotope analysis of charred archaeological grains provides a remarkably precise scale of information: the growing conditions under which a plant was cultivated in a single field and season. Here we investigate how the measurement of single individual grains or aggregate bulk samples for carbon and nitrogen isotopes impacts how we characterize variation and, consequently, our interpretations of ancient cultivation practices. Using experimentally grown barley (Hordeum vulgare var. nudum), this work investigates {delta}13C and {delta}15N intra-panicle variation between both uncharred and charred individual grains from four plants. We found limited intra- and inter-panicle isotopic variation in single grain isotope values, ca. 0.5{per thousand} in {delta}13C and ca. 1{per thousand} in {delta}15N, reemphasizing the degree to which grains are representative of their local growing conditions. To explore the interpretive impact of aggregate versus single-grain isotopic sampling, we measured charred barley recovered from a single storage context excavated from Trench 42 (ca. 1900 BCE) at Harappa. Aggregate samples of a random selection of Trench 42 barley demonstrated remarkable inter-sample homogeneity, with a less than 0.5{per thousand} difference in {delta}13C and {delta}15N values, reinforcing the ability of aggregate samples to capture a representative isotopic average of a single depositional context. However, the measurement of single-grains revealed moderate 2-3{per thousand} variation in {delta}13C, and an outstandingly wide isotopic variation of ca. 8{per thousand} in {delta}15N values, indicating the degree to which cultivation practices varied beyond what the bulk samples indicated. These results highlight how decisions in the selection and measurement of archaeological grains for isotopic analysis impact data resolution, with profound consequences for understanding past agricultural diversity.