Isotopic evidence for anthropogenic lead exposure on a 17th/18th century Barbadian plantation

Abstract Objectives To identify and characterize anthropogenic lead sources on a 17th/18th century Barbadian plantation and to test if lead isotope analyses can be used to identify the geographic origins of first‐generation African captives. Materials and Methods We carried out lead (Pb) isotope analyses on dental enamel samples from 24 individuals from the Newton Plantation Cemetery in Barbados, which had previously been analyzed for strontium (Sr) and oxygen (O) isotope composition (Schroeder et al., American Journal of Physical Anthropology, 2009, 139:547–557) and Pb concentrations (Schroeder et al., American Journal of Physical Anthropology, 2013, 150:203–209. Results We are able to identify British Pb sources, and more specifically Bristol/Mendips Pb, as the most likely source of anthropogenic Pb on the plantation, highlighting the impact of the British Atlantic economy on the lives of enslaved peoples in Barbados during the period of plantation slavery. Furthermore, we find that there is only one clear outlier among seven individuals who had previously been identified as African‐born based on their enamel Sr isotope composition (Schroeder et al., American Journal of Physical Anthropology, 2009, 139:547–557). All other individuals present a very homogenous Pb isotope composition, which overlaps with that of British Pb sources. Conclusion Our results indicate that while Pb isotope analyses can help identify and further characterize the sources of anthropogenic Pb in plantation settings, they might not be suited for identifying the origins of African‐born individuals in diasporic contexts.

enslaved populations comes from the Newton Plantation cemetery in Barbados (Figure 1), where 17th-18th century skeletal remains attest to frequent anthropogenic exposure. Lead toxicity can result from ingestion or inhalation of vapors, and colonial accounts provide widespread evidence of gastrointestinal illness (constipation, nausea, and loss of appetite), infertility, and neurosensory complications (lethargy, seizures, and weakness), which are clinically documented in cases of lead intoxication (Handler et al., 1986).
Various common household items were sources of lead contamination, including earthenwares, pewter drinking and serving vessels, flour, cosmetics, hair dyes, house paints, and medicaments. Water sources could become contaminated through lead-lined cisterns and gutters, though less regularly available to the enslaved who collected water from nearby ponds (Handler et al., 1986). The most likely routes of exposure for the enslaved were, arguably, rum and molasses, which were contaminated through leaded fixtures and piping (Handler et al., 1986). Average bone lead content from Newton (n = 48; 117.6 ± 94.9 ppm or μg lead/g bone ash) was found to be nearly four times greater than reported for enslaved groups from the United States. (Aufderheide et al., 1985;Corruccini et al., 1987;Handler, 2006;Handler et al., 1986), and the lead levels of adults of both sexes appear to have resulted from chronic exposure throughout life (Corruccini et al., 1987). Schroeder et al. (2013) subsequently evaluated dental enamel in 26 individuals from the site, which, with earlier bone lead data, showed a clear pattern of chronic lead accumulation beginning at, or even before, birth.
Exposure during infancy and early childhood, as indicated by dental enamel, may have occurred in utero through endogenous maternal bone stores released into circulating blood, crossing the placenta (Ronchetti et al., 2006;Winecker, Ropero-Miller, Broussard, & Hammett-Stabler, 2002). However, the exceptionally high levels of lead observed in perinates are unlikely to have resulted solely from gestational release alone. Lead contamination may also have resulted from lactation (Gulson et al., 1998) and perhaps also from pica/geophagy, which is the craving and consumption of nonfood items: a practice that was documented archivally for the enslaved on the island (Handler, 2006) and a common route of contamination in children in industrialized nations today (Choi, Tanaka, Koren, & Ito, 2008).

| Lead isotope analysis of skeletal remains for provenance research
Over the last several decades, there has been a substantial increase in the application of isotopic approaches for tracking modern and ancient human and animal movements (Makarewicz & Sealy, 2015).
Isotope analyses of archeological skeletal remains for paleomobility and provenance studies have primarily focused on strontium (Sr) or oxygen (O) isotope systems. A recent trend has been the increasing use of multi-isotope proxy data including both 87 Sr/ 86 Sr and δ 18 O, sometimes in conjunction with carbon (C) and nitrogen (N) isotope analyses. This trend reflects the widely acknowledged observation that there are inherent limitations of single isotope proxy data. One of the more confounding limitations is the issue of equifinality (Price, Burton, & Stoltman, 2007), whereby multiple locations possess similar isotope ranges making it impossible to distinguish between two or more possible places of origin. Owing to these persistent issues, increasing the number of analyzed isotopes may provide more reliable identifications of local versus nonlocal origins and potentially improve the accuracy of assessments of geographic place of origin (Laffoon et al., 2017). Lead (Pb) isotope analysis of human dental enamel arguably offers the greatest potential for contributing to paleomobility research owing to its high spatial variability and well documented geological and geographical constraints (Evans, Pashley, Madgwick, Neil, & Chenery, 2018).
The principles of the Pb isotope approach are comparable to the Sr isotope method (Bentley, 2006;Ericson, 1985;Price, Burton, & Bentley, 2002). Lead from the environment is ingested or inhaled and becomes incorporated into body tissues (Gulson, 2012;Kamenov & Gulson, 2014). It tends to be fairly quickly removed from soft tissues whereas the skeletal system holds the greatest proportion of the overall Pb load as Pb replaces Ca within bioapatite, the primary mineral component of teeth and bones (Gulson, 2012). Dental enamel, unlike bone and dentine, does not undergo subsequent remodeling, and therefore it preserves the isotope signal of the biochemical environment where it formed (Budd, Millard, Chenery, Lucy, & Roberts, 2004;Gulson & Gillings, 1997). As such, archaeological enamel Pb signatures can serve as geographic proxy indicators and provide F I G U R E 1 Map of Barbados showing the location of the Newton Plantation information concerning an individual's childhood or natal origins (Budd et al., 2004;Carlson, 1996;Chiaradia, Gallay, & Todt, 2003;Montgomery, Evans, Powlesland, & Roberts, 2005).
There are four naturally occurring isotopes of Pb, of which 204 Pb is not radiogenically produced whereas the other three isotopes 206 Pb,207 Pb,and 208 Pb are produced by the radioactive decay of other elements ( 238 U, 235 U, and 232 Th, respectively) and, thus, are widely used for geochronology applications (Faure & Mensing, 2005).
The relative abundance of the Pb isotopes to each other (Pb isotope ratios) in rocks and minerals is dependent on the original amounts of Pb, U, and Th in the parent material and its age. As such, geographic variation in natural environmental Pb isotope composition is primarily conditioned by geological variation (age and lithology of bedrock and sediments).

| Materials
We analyzed dental enamel samples from 24 individuals from the Newton Plantation skeletal assemblage, one of the most well-documented and well-studied skeletal assemblages dating to the era of plantation slavery in the Caribbean (Handler & Corruccini, 1983;Handler & Lange, 1978). Dental samples selected for Pb isotope analyses (n = 24) mostly comprise first molars (M1) and were previously analyzed for strontium and oxygen isotope composition (Schroeder et al., 2009) and Pb concentrations

| Analytical procedures
Each sample was rinsed several times with distilled water with 5 min of ultrasonication between each step. The samples were then dissolved in 2 M HCl, spiked with a 205 Pb-235 U tracer solution of known concentration, dried down and redissolved in 1 M HBr. Lead from the samples was purified with a standard HBr-HNO 3 column chemistry using Eichrom anion resin, with each sample processed twice on the same column (Connelly & Bizzarro, 2009).
The purified Pb was loaded onto an outgassed zone-refined Re ribbon with Si-gel and H 3 PO 4 and analyzed using a ThermoScientific™ Thermal Ionizing Mass Spectrometer equipped with nine Faraday detectors and one axial secondary electron multiplier-ion counting (SEM-IC) system. The natural isotopes of Pb were measured in static mode on Faraday detectors, whereas 205 Pb was measured in the SEM-IC. The Faraday-SEM-IC conversion factor was determined at the start of the run (after focusing the filament) by peak jumping 204 Pb between the SEM-IC and the Faraday cup H1. The 206 Pb/ 205 Pb ratio was used to determine the amount of Pb in each sample. The small contribution of natural Pb isotopes from the tracer to the sample is corrected according to the amount of 205 Pb present in the analyses. Long-term instrument performance is monitored by running the NIST981 Pb standard at the start and end of each measurement session. Blank contributions for this chemical procedure and loading the filaments are less than 0.5 pg, which is trivial given the amount of Pb from each sample.

| RESULTS
Relevant sampling information and Pb isotope results are listed in Table 1 and displayed in Figures 2-5. Similar to previously reported results , the newly generated Pb concentrations are highly variable ranging from~0.5 to 40 ppm (Table 2)  higher Pb concentrations. This pattern of inversely correlated Pb concentrations and Pb isotope ratios is consistent with the long-term trend in human enamel Pb data in Europe that has been termed cultural focusing (Montgomery et al., 2005; Figure 3) and suggests that samples from Newton Plantation with higher Pb concentrations and similar Pb isotope ratios are the result of (childhood) exposure to anthropogenic Pb sources .
All 24 samples possess Pb isotope compositions that are distinct from those reported for Barbadian geological materials ( Figure 4) and do not fall on the trendline for modern North American (anthropogenic) Pb sources (Kamenov & Gulson, 2014). Overall, the Pb isotope compositions of the entire data set are fairly homogeneous (Table 2) with the exception of one extreme outlier (N52). Individual N52 had F I G U R E 4 Comparison of Pb isotope data from locals and nonlocals at Newton Plantation, Barbados compared to geological Pb isotope data from Barbados, Britain, and West Africa. Data for Newton Plantation (this study); for Barbados basement rocks (Carpentier, Chauvel, & Mattielli, 2008); for British lead-ore minerals (Fletcher, Swainbank, & Colman, 1993;Haggerty, Budd, Rohl, & Gale, 1996;Rohl, 1996 However, six out of the seven nonlocals possess Pb isotope ratios that are indistinguishable from the local population, and by extension also fall within the range of British Pb sources. There are several, nonmutually exclusive mechanisms that may account for these observed patterns in the data. First, this could simply result from overlap in the ranges of Pb isotope variation (Figure 4) between Pb sources in Britain (reflected in the enamel Pb isotope ratios of the local population) and in West Africa (reflected in the enamel Pb isotope ratios of six of the nonlocals). Unfortunately, it is not possible to further constrain the potential origins of these individuals on the basis of the Pb isotope data alone, as the spatial variation of Pb isotopes in Africa is not well characterized. To our knowledge, little or no human Pb isotope data from Africa are available at this time for comparative analysis. However, given the documented geological complexity of West Africa, we would expect a high degree of spatial variation in both geological and bioavailable Pb isotopes in this region of the world. As such, it is not likely that the six nonlocal (African-born) individuals would possess highly diverse geographic origins evidenced by their heterogeneous strontium isotope ratios and yet simultaneously also have nearly identical enamel Pb isotope ratios.
Second, it is possible given the long history of interaction and colonialism that West Africans (i.e., born and raised in Africa) in the 17th-18th century could have been exposed to technological (anthropogenic) lead imported or traded from a European source, although there is no direct evidence that this applies to the first-generation Africans identified at Newton.
Third, the homogenous Pb isotope ratios in the Newton nonlocals reflect contamination from anthropogenic Pb sources on Barbados. As we have found no reported cases of in vivo Pb contamination of fully mature dental enamel in the biomedical literature, postmortem contamination (diagenesis) seems to be the most F I G U R E 5 Enamel 208 Pb/ 206 Pb versus 207 Pb/ 206 Pb isotope ratios from Newton Plantation, Barbados, compared to geological Pb-ore variation in Britain (Farmer, Eades, & Graham, 1999;Fletcher et al., 1993;Haggerty et al., 1996;Rohl, 1996;Scaife et al., 2001;Shepherd et al., 2009)  parsimonious explanation. In fact, previous research has identified postmortem alteration of Pb isotope ratios in archaeological skeletal remains in diverse geographical and cultural contexts (Chiaradia et al., 2003;Dudás et al., 2016) including in the Caribbean region (Giovas et al., 2016). Perhaps counter-intuitively, individuals with the lowest Pb concentrations in their dental enamel (purportedly reflecting geogenic, as opposed to anthropogenic, Pb sources) are the most susceptible to alteration of their Pb isotope compositions (Kamenov et al., 2018). This process is illustrated with a simple mixing model ( Figure 6) demonstrating that even minor contributions of non-biogenic Pb can significantly alter the Pb isotope ratios measured in archeological dental enamel. This contrasts sharply with strontium with typical contributions in human skeletal remains (~200-500 ppm) that are several orders of magnitude greater than the enamel Pb concentrations observed in the Newton nonlocal data set (0.4-2.7 ppm). The one nonlocal individual with a distinct Pb isotope composition may have preserved an authentic biogenic (African) Pb signal in her dental enamel. The exact mechanisms accounting for the observed variation in Pb contamination are not known but previous research has implicated multiple variables such as burial depth, burial soil chemistry, and soil particle size (e.g., Chiaradia et al., 2003;Giovas et al., 2016).
Overall, we conclude that Pb isotope measurements are not very useful for human provenance studies for the Newton population in particular and that considerable caution is merited in the application of Pb isotope analysis in other early modern contexts given the potential susceptibility of dental enamel with low Pb concentrations to contamination with modern or past anthropogenic Pb. Nevertheless, the finding that the locally born fraction of the Newton population possesses both high Pb concentrations and enamel Pb isotope ratios that match the Pb isotope composition of British (Bristol/ Mendips) Pb ore sources is still valid and most likely reflects exposure to high levels of anthropogenic Pb from early childhood at the Newton Plantation, with devastating health consequences for the population.
The health impacts of lead were common though poorly understood in British colonies in North America and the West Indies during the 17th and 18th centuries (Aufderheide et al., 1985;Corruccini et al., 1987;Handler et al., 1986;Schroeder et al., 2013).
Physicians frequently described in the medical literature a mysterious illness that was variably referenced by such terms as "dry bellyache," "dry gripes," "West Indian dry gripes," and "nervous colick," among others (Handler et al., 1986, p. 408). Characterized by debilitating abdominal cramps, severe constipation without diarrhea, and weakness or paralysis of the limbs, this illness was widely known to be endemic among the white populations and descriptions are consistent with the clinical manifestations of lead poisoning (Handler et al., 1986).
Although the illness was very rarely mentioned for African populations, Handler and colleagues provide strong inferential evidence that the enslaved also suffered from the effects of lead that went unrecognized by early West Indian physicians (Corruccini et al., 1987;Handler et al., 1986). Ingestion of food tainted by imported lead is widely implicated in cases of lead poisoning on plantations throughout the British colonies (Aufderheide et al., 1985;Aufderheide, Neiman, Wittmers, & Rapp, 1981;Wittmers, Aufderheide, Rapp, & Alich, 2002), though Handler and colleagues have argued that the enslaved would have low access to many household items such as pewter, powders, flours and other sources that commonly sickened white households and domestic servants.
Rather, they implicated rum distillation and consumption as the most likely sources of lead on Caribbean plantations, notably on Barbados, where skeletal and dental evidence has been used to attest to high lead levels in enslaved rum producers (Corruccini et al., 1987;Handler et al., 1986;Schroeder et al., 2013). The high to extreme lead values observed at Newton during infancy to early childhood, along with near absence of developmental features that were predicted with rum consumption, seem to suggest greater complexity in routes of exposure

| CONCLUDING REMARKS
This study presents the results of Pb isotope analyses of human dental enamel from the 17th/18th century enslaved African population of Newton Plantation, Barbados. The Pb isotope data are interpreted in the context of archaeological, bioarchaeological, historical, and previously generated strontium isotopic data .
In contrast to expectations, the Pb isotope data neither mirrored nor