Rare Earth Elements are important proxies for tracing the evolution and redox history of the Earth's hydrosphere. Many biogenic and abiogenic archives have been analyzed for their REE contents with qualified success, which may be due to depositional and post-depositional alteration effects. Review of the database reveals a lack of a rigorous sample processing protocol and probably the leading cause for the spurious REE results reported in the literature.We propose a ‘sample cleaning protocol’ that should satisfy the most stringent demands for procuring reliable and robust REEs from marine materials. Without cleaning, the results may represent REE compositions not only of the shell's structure but also of lattice-bound oxides, detritus, particulates and organic remnants. Thus, cleaning of material is a fundamental step that should be conducted with care and attention to detail prior to analyzing their REE contents. To achieve this goal, valves from recently dead Liothyrella neozelanica recovered from deep water of the South Pacific Ocean (north of New Zealand) were subjected to five different cleaning procedures. In Procedure-1 (P-1) valve fragments were only washed and rinsed with distilled water. In Procedure-2 (P-2) a set of fragments was immersed in 2.5% hydrogen peroxide (H2O2) for three continuous days and then water washed. In Procedure-3 (P-3) valve fragments were physically cleaned using a sharp stainless-steel blade and then water washed. In Procedure-4 (P-4) a set of fragments was physically cleaned then immersed briefly in 10% hydrochloric acid (HCl) until they were deemed clean, and then water washed. In Procedure-5 (P-5) the last set of fragments was processed using all cleaning protocols such as physical scraping, H2O2, HCl leaches and water washing.Detritus and nano-particulates adsorbed on the calcitic structure of brachiopod shells including the proteinaceous periostracum may lead to elevated ∑REE content, anomalous Ce/Ce* and elevated Mn, Fe and U concentrations as documented by the P-1 protocol results. Hydrogen peroxide immersion (P-2) eliminates primarily the organic tissue, which leads to an invariant Ce/Ce* anomaly. Physical cleaning (P-3) removes adsorptive nano-particulates and the periostracum, and leads to depleted ∑REE and lower Mn, Fe, and U concentrations with a slight Ce/Ce* anomaly. Physical cleaning followed by chemical cleaning (P-4 and P-5) removes adsorptive particulates, organic remnants, the periostracum and the primary layer. These last two procedures produce drastic reductions in ∑REE, Mn, Fe, and U concentrations and normal Ce/Ce* anomalies with typical seawater REE signatures in brachiopod calcite.Procedure P-4 was tested on Silurian, Pennsylvanian and Permian brachiopods and enclosing whole rock as well as conodonts and fish debris to assess the reliability of REE contents in Deep-Time fossils and carbonates. The brachiopod shells contain ∑REEs, Mn, Fe, and U concentrations and Ce/Ce* values comparable to their modern counterparts, while in the coeval micritic whole rock they are slightly to significantly enriched. Those with elevated Fe and Mn (by more than 300 and 100ppm, respectively) their Ce/Ce* values should be adjusted by −13% if they are to be considered in paleoredox investigations. In contrast, REE contents of conodonts give the typical ‘bell shape’ trend for the lanthanides indicative of post-depositional alteration coupled with extraneous Ce/Ce* anomaly values. Strict adherence to the proposed sample processing protocol is critical if we want reliable and robust ‘seawater’ REE signatures from brachiopods and whole rocks, and potentially other archives.