Evidence in the literature suggests that the trace element chromium may have a role in glucose homeostasis through the regulation of insulin action. We have previously reported a significant reduction in plasma chromium levels in healthy individuals, following a 75 g oral glucose load, and after meals and glucose-dependent uptake of chromium in insulin-dependent tissues in vitro. However, in vivo it is unclear whether the changes in plasma chromium relate to changes in plasma glucose or insulin. The present study describes a series of euglycaemic hyperinsulinaemic clamps designed to attempt to define the initiator of changes in plasma chromium levels in ten healthy individuals. The data showed a significant (P < 0.01) reduction in fasting plasma chromium levels following glucose infusion and an initial bolus of insulin. Significant (P < 0.02) increases in post-clamp urinary chromium excretion were insufficient to explain the decrease in plasma levels. During the recovery phase of an extended two-phase clamp protocol we found plasma insulin levels decreased by 70% within 10 min, associated with an increase in plasma chromium levels of 30% and no significant change in plasma glucose level. These data indicate that alterations in plasma glucose are unlikely to be directly related to changes in plasma chromium, whilst supporting the hypothesis that plasma insulin may influence plasma levels of this trace element. In contrast, plasma zinc was unaffected throughout these clamp studies.