Effect of sample preparation techniques for carbon and nitrogen stable isotope analysis of hydroxyapatite structures in the form of elasmobranch vertebral centra

RATIONALE: Bulk stable isotope analysis (SIA) provides an important tool for the study of animal ecology. Elasmobranch vertebral centra can be serially sampled to obtain an isotopic history of an individual over ontogeny. The measured total δ(13)C value, however, may be misinterpreted due to the inclusion of the (13)C-rich inorganic portion. Hydrochloric acid (HCl) is commonly used to remove the inorganic portion of hydroxyapatite structures before undertaking SIA, but more recently ethylenediaminetetraacetic acid (EDTA) has been recommended for elasmobranch vertebrae. These acid treatments may introduce uncertainty on measured δ(13)C and δ(15)N values above instrument precision and the effect of small sample size remains untested for elasmobranch vertebrae. METHODS: Using a non-dilution program on an isotope ratio mass spectrometer the minimum sample weight of vertebrae required to obtain accurate isotopic values was determined for three shark species: white (Carcharodon carcharias), tiger (Galeocerdo cuvier), and sand tiger (Carcharias taurus). To examine if acid treatment completely removes the inorganic component of the vertebrae or whether the technique introduces its own uncertainty on measured δ(13)C and δ(15)N values, vertebrae samples were analyzed untreated and following EDTA treatment. RESULTS: The minimum sample weight required for accurate stable isotope values and the percentage sample yield following EDTA treatment varied within and among species. After EDTA treatment, white shark vertebrae were all enriched in (13)C and depleted in (15) N, tiger shark vertebrae showed both enrichment and depletion of (13)C and (15)N, and sand tiger shark vertebrae were all depleted in (13)C and (15)N. CONCLUSIONS: EDTA treatment of elasmobranch vertebrae produces unpredictable effects (i.e. non-linear and non-correctable) among species in both the percentage sample yield and the measured δ(13)C and δ(15)N values. Prior to initiating a large-scale study, we strongly recommend investigating (i) the minimum weight of vertebral material required to obtain consistent isotopic values and (ii) the effects of EDTA treatment, specific to the study species and the isotope ratio mass spectrometer employed.