Résumé : Skeletal Mg/Ca in echinoderms is thought to be related to ambient seawater Mg2+/Ca2+ ratio, which prompts the use of well-preserved fossil echinoderms as paleoseawater Mg2+/Ca2+ proxies. Nonetheless, experimental studies testing the effect of seawater Mg2+/Ca2+ ratio on echinoderm skeleton are limited to one echinoderm class, i.e., sea urchins (echinoids). Here, we investigated the effect of decreased seawater Mg2+/Ca2+ ratio on skeletal composition (Mg/Ca ratio) of a sea star Asterias rubens and a brittle star Ophiocomina nigra. Specimens were tagged with manganese and then cultured under three different Mg2+/Ca2+ molar ratios (~5.2, ~2.5, ~1.5 mol/mol) in order to simulate variations of ambient seawater Mg2+/Ca2+ ratio that existed throughout the Phanerozoic. Decreased Mg2+/Ca2+ in seawater did not affect the respiration rates or inhibit calcification of sea stars, which suggests that they were not significantly stressed by the treatment. It did, however, clearly affect growth and mortality in brittle stars, for which reliable geochemical data could not be obtained. Under decreased seawater Mg2+/Ca2+ sea stars produced a skeleton with decreased Mg/Ca ratio. This is consistent with the hypothesis that the skeletal chemistry in echinoderms is influenced by seawater chemistry and, consequently, that fossil echinoderms may preserve a record of paleoseawater Mg2+/Ca2+. Although our data from well-preserved fossil sea stars from the Middle Jurassic and Miocene are consistent with low Mg2+/Ca2+ ratio of the Jurassic “calcite sea” and high Mg2+/Ca2+ ratio of the Miocene "aragonite sea", the accuracy of paleoseawater Mg2+/Ca2+ reconstructions from fossil echinoderms is limited by the fact that Mg partition coefficients vary significantly between different echinoderm species.