Living and sub-recent ostracods were collected from four lakes and foul springs, located in Belgium, France and Spain, characterized by low water Mg/Ca ratios (from 0.08 to 0.65). Valves of ostracods belonging to Cyprideis torosa, Candona neglecta, C. marchica and C. candida have been extensively analysed for their magnesium and strontium contents with respect to calcium. Comparisons of these results are made using the same elements from the waters in which the ostracod valves calcified. Our results offer new and somewhat different interpretations with respect to the uptake of Mg and Sr in the calcitic ostracod valves. These findings include the following. (1) The partition coefficient L)(Sr for ostracods belonging to the same genus (Candona) is not a constant. The observed variability in D(Sr) values could be due to the rare of biological calcification, although more investigations are necessary to assess the influence of this factor; it is possible that some taxa grouped under the genus Candona belong to a different genus. (2) Even at the species level (Cyprideis torosa), the ostracod D(Sr) cannot be considered to be constant. At low water Mg/Ca ratio (0.65), our D(Sr) value for C. torosa is higher than those determined by others for C. torosa calcified in evolved marine water. (3) The Mg/Ca ratio of the water appears as a major factor influencing D(Mg) for ostracods. For low water Mg/Ca values (< similar to 2), D(Mg) drastically increases. This phenomenon is also effective for inorganic calcite; implying a non-biologic effect that could be linked to Mg2+ adsorption. Consequently, large variations in ME content of fossil ostracod valves from selected lacustrine sequences, with expected low water Mg/Ca(< similar to 2), can be due to small variations in water Mg/Ca. (C) 1998 Elsevier Science B.V. All rights reserved.
Wansard, G., De Deckker, P., & Julia, R. (1998). Variability in ostracod partition coefficients D(Sr) and D(Mg) implications for lacustrine palaeoenvironmental reconstructions. Chemical Geology, 146(1-2), 39-54. https://doi.org/10.1016/S0009-2541(97)00165-4 (Original work published 1998)