The toxicity of metals and metalloids is well known to depend strongly on speciation in the environment. This is particularly evident in the case of chromium. Chromium (III) is generally an essential nutrient (at low concentrations), whilst the Cr(VI) oxidation state is highly toxic and carcinogenic in humans. Whilst Cr(VI) is well recognized to be highly toxic to a range biota, there is relatively little available data enabling modelling of the terrestrial ecotoxicity of Cr as influenced by environmental factors, such as pH and ionic compositoin of pore-waters. The absorption and toxicity of metallic cations and anions are influenced by the presence of potentially competiting ions, such as calcium and phosphate. A series of models have been proposed accounting for such interactions in the bathing solution around biological membranes. These models include the windermere humic acid model (WHAM), electrostatic and the biotic ligand models (BLM). These models deal almost exclusively with cationic metals such as nickel, copper and zinc and have exluded metallic anions such as arsenate and chromate. In the case of Cr(III), BLM models have been reported in plants and could also potentially be modelled using WHAM. However, the more environmentally concerning Cr(VI) has scarcely been investigated, both in terms of soil solution impacts on phytotoxicity, but also, the development of BLM modelling parameters. In this study, we developed a terrestrial BLM-like model for chromate in soil solution using 10 contrasting soils with a range of pH, bicarbonate, sulphate and phosphate concentrations using Cucumis sativa L.. Plants were exposed to soils with increasing Cr(VI) using a 28 day assay. Soil solution was sampled with Rhizon samplers and chromium speciation determined using liquid chromatography inductively coupled plasma mass spectrometry. The results showed that unlike previously reported studies, there was no clear interaction with sulphate or phosphate with chromate toxicity. However, there was a clear interaction between OH^- and HCO₃^-_. Therefore, we successfully developed a BLM-like model for chromate which was a function of OH^- and HCO₃^- activities (a root mean squre error of 19.4%). The nature of the interaction between chromate and OH^-/HCO₃^- has implications for intepretation of BLM models, particularly relating to other metallic anions.