No evidence for effects of mill-grinding on δ13 C, δ15 N and δ34 S values in different marine taxa.

Abstract

The analysis of natural variation in light stable isotopes such as carbon (δ13 C), nitrogen (δ15 N) and sulfur (δ34 S) plays an important role in deepening our understanding of ecosystems. To avoid misinterpretation, robust results are required, where pre-treatment steps such as sample homogenization are crucial to guarantee representative samples. The grinding of samples using stainless steel balls in polypropylene tubes (e.g., laboratory bead-beater) has been identified as a potential source of bias. We tested possible effects of mill-grinding (e.g., contamination) of samples of coastal marine taxa including primary producers, primary consumers and higher trophic level fish. We compared potential impacts of homogenization by mill-grinding with hand-grinding over an extended time on δ13 C, δ15 N and δ34 S values. One-way Welch's analysis of variance (ANOVA) showed that there were no statistical differences between methods for all the studied taxa. Also, repeated measures ANOVA showed no evidence of effects of grinding for extended times (from 30 to 120 seconds) for δ13 C, δ15 N and δ34 S values. We found no evidence that grinding samples in polypropylene tubes in a bead-beater resulted in any marked alteration of the isotopic composition on the studied samples, e.g., through contamination by plastic. As such, we consider mill-grinding as an appropriate method for the homogenization of samples from a range of different marine taxa, which under controlled conditions did not affect δ13 C, δ15 N and δ34 S analysis.