Phylogeochemistry: exploring evolutionary constraints on belemnite rostrum element composition

Abstract. The biogenic carbonate hardparts of a large range of marine organisms are the most important geochemical archives of Earth’s climate dynamics through time and the evolution of life. That said, biomineralisation pathways, i.e., the secretion of mineral phases by organisms, are complex and may differ significantly between different taxa. In light of this, it is critically important to evaluate if related taxa might display similar hard parts geochemistry. If so, this relation might bear information on evolutionary relationships and has great significance in carbonate archive research. Here, we test the evolutionary constraints on main and trace element ratios of belemnite rostra using Bayesian phylogenetic tools. For this purpose, we assembled a large dataset on element ratios from 2241 published samples of belemnite rostra and used comparative Bayesian phylogenetic tools to reconstruct ancestral states and evolutionary rates. While Mn/Ca and Fe/Ca appear to be taxon-independent and probably mainly reflect environmental and diagenetic effects, Mg/Ca and Sr/Ca display stronger taxon-specific patterns, even though their interpretation remains complex. The evolutionary rates are high, with average estimated changes in element ratios of 12.4 % (Mg/Ca) and 12.3 % (Sr/Ca) of the overall mean element ratio per 1 million years. While the distribution of Sr/Ca ratios is relatively homogeneous across the tree, Mg concentrations are divided among two distinct groups (< 5.5 and >7.5 mmol/mol, respectively), with at least five evolutionary transitions between them. Beyond carbonate archive research, our phylogenetic analysis provides insights into the evolution of belemnites. This study highlights the complex interplay between evolutionary, ontogenetic, environmental and diagenetic effects and calls for caution when using belemnite element ratios as proxies for palaeoclimatic studies. We propose the term ’phylogeochemistry’ for the investigation of geochemical data using phylogenetic modelling techniques.