Abstract
Skeletal remains in archaeological strata are often assumed to be of similar ages. Here we show that combined Sr and O isotope analyses can serve as a powerful tool for assessing fish provenance and even for identifying fossil fish teeth in archaeological contexts. For this purpose, we established a reference Sr and O isotope dataset of extant fish teeth from major water bodies in the Southern Levant. Fossil shark teeth were identified within Iron Age cultural layers dating to 8–9th century BCE in the City of David, Jerusalem, although the reason for their presence remains unclear. Their enameloid 87Sr/86Sr and δ18OPO4 values [0.7075 ± 0.0001 (1 SD, n = 7) and 19.6 ± 0.9‰ (1 SD, n = 6), respectively], are both much lower than values typical for modern marine sharks from the Mediterranean Sea [0.7092 and 22.5–24.6‰ (n = 2), respectively]. The sharks’ 87Sr/86Sr are also lower than those of rain- and groundwater as well as the main soil types in central Israel (≥0.7079). This indicates that these fossil sharks incorporated Sr (87Sr/86Sr ≈ 0.7075) from a marine habitat with values typical for Late Cretaceous seawater. This scenario is in line with the low shark enameloid δ18OPO4 values reflecting tooth formation in the warm tropical seawater of the Tethys Ocean. Age estimates using 87Sr/86Sr stratigraphy place these fossil shark teeth at around 80-million-years-old. This was further supported by their taxonomy and the high dentine apatite crystallinity, low organic carbon, high U and Nd contents, characteristics that are typical for fossil specimens, and different from those of archaeological Gilthead seabream (Sparus aurata) teeth from the same cultural layers and another Chalcolithic site (Gilat). Chalcolithic and Iron Age seabream enameloid has seawater-like 87Sr/86Sr of 0.7091 ± 0.0001 (1 SD, n = 6), as expected for modern marine fish. Fossil shark and archaeological Gilthead seabream teeth both preserve original, distinct enameloid 87Sr/86Sr and δ18OPO4 signatures reflecting their different aquatic habitats. Fifty percent of the analysed Gilthead seabream teeth derive from hypersaline seawater, indicating that these seabreams were exported from the hypersaline Bardawil Lagoon in Sinai (Egypt) to the Southern Levant since the Iron Age period and possibly even earlier.
Highlights
Past fish habitats have traditionally been reconstructed based on taxonomic identification of fish remains recovered in archaeological sites (e.g., Wheeler and Jones, 1989; Zohar and Biton, 2011; Zohar, 2017)
The recovery of a relatively large number of shark teeth in the Iron Age assemblage of the City of David together with diverse fish remains is puzzling for two main reasons: (1) shark teeth are rare in archaeological sites in the Southern Levant (Table 1); (2) this study clearly demonstrates that these shark teeth are fossils and do not represent shark consumption
Multiple Late Cretaceous (80.3 ± 3.2 millions of years (Ma)) fossil shark teeth were encountered in the same Iron Age cultural layers of the City of David, Jerusalem together with a wide diversity of archaeological fish originating from the Mediterranean Sea and the Nile
Summary
Past fish habitats have traditionally been reconstructed based on taxonomic identification of fish remains recovered in archaeological sites (e.g., Wheeler and Jones, 1989; Zohar and Biton, 2011; Zohar, 2017). The strontium isotope composition (87Sr/86Sr) of the wellpreserved bioapatite of fish tooth enameloid has been widely used as a chemostratigraphic dating method for marine sediments (e.g., Ingram, 1995; Kocsis et al, 2009; Harrell et al, 2016) as a proxy to infer the palaeosalinity levels of oceanic basins and brackish water bodies (e.g., Schmitz et al, 1991, 1997; Bryant et al, 1995; Reinhardt et al, 1998; Kocsis et al, 2009), as well as for tracing past freshwater habitat use in sharks (e.g., Fischer et al, 2013) and to track their migration from the marine realm into fresh water systems (Kocsis et al, 2007, 2015). The 87Sr/86Sr in fish teeth can be used as a proxy to distinguish between freshwater and marine habitats (Figure 1A; e.g., Schmitz et al, 1997; Kocsis et al, 2007, 2014; Tütken et al, 2011; Fischer et al, 2013), whereas for brackish water bodies the 87Sr/86Sr depends on the sea-to-freshwater mixing ratio (Bryant et al, 1995; Reinhardt et al, 1998)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
