The epidemiological association between black cherry trees and mare reproductive loss syndrome has focused attention on cyanide and environmental cyanogens. This article describes the toxicokinetics of cyanide in horses and the relationships between blood cyanide concentrations and potentially adverse responses to cyanide. To identify safe and humane blood concentration limits for cyanide experiments, mares were infused with increasing doses (1-12 mg/min) of sodium cyanide for 1 h. Infusion at 12 mg/min produced clinical signs of cyanide toxicity at 38 min; these signs included increased heart rate, weakness, lack of coordination, loss of muscle tone, and respiratory and behavioral distress. Peak blood cyanide concentrations were about 2500 ng/mL; the clinical and biochemical signs of distress reversed when infusion stopped. Four horses were infused with 1 mg/min of sodium cyanide for 1 h to evaluate the distribution and elimination kinetics of cyanide. Blood cyanide concentrations peaked at 1160 ng/mL and then declined rapidly, suggesting a two-compartment, open model. The distribution (alpha) phase half-life was 0.74 h, the terminal (beta phase) half-life was 16.16 h. The mean residence time was 12.4 h, the steady-state volume of distribution was 2.21 L/kg, and the mean systemic clearance was 0.182 L/h/kg. Partitioning studies showed that blood cyanide was about 98.5% associated with the red cell fraction. No clinical signs of cyanide intoxication or distress were observed during these infusion experiments. Mandelonitrile was next administered orally at 3 mg/kg to four horses. Cyanide was rapidly available from the orally administered mandelonitrile and the C max blood concentration of 1857 ng/mL was observed at 3 min after dosing; thereafter, blood cyanide again declined rapidly, reaching 100 ng/mL by 4 h postadministration. The mean oral bioavailability of cyanide from mandelonitrile was 57% +/- 6.5 (SEM), and its apparent terminal half-life was 13 h +/- 3 (SEM). No clinical signs of cyanide intoxication or distress were observed during these experiments. These data show that during acute exposure to higher doses of cyanide (~600 mg/horse; 2500 ng/mL of cyanide in blood), redistribution of cyanide rapidly terminated the acute toxic responses. Similarly, mandelonitrile rapidly delivered its cyanide content, and acute cyanide intoxications following mandelonitrile administration can also be terminated by redistribution. Rapid termination of cyanide intoxication by redistribution is consistent with and explains many of the clinical and biochemical characteristics of acute, high-dose cyanide toxicity. On the other hand, at lower concentrations (<100 ng/mL in blood), metabolic transformation of cyanide is likely the dominant mechanism of termination of action. This process is slow, with terminal half-lives ranging from 12-16 hours. The large volume of distribution and the long terminal-phase-elimination half-life of cyanide suggest different mechanisms for toxicities and termination of toxicities associated with low-level exposure to cyanide. If environmental exposure to cyanide is a factor in the cause of MRLS, then it is likely in the more subtle effects of low concentrations of cyanide on specific metabolic processes that the associations will be found.