To better understand the pharmacology of the thiopentone enantiomers, we studied their quantitative electroencephalographic effects and their distribution into vital tissues. 2. Adult Wistar rats were infused with rac-, R- or S-thiopentone at 4 mg kg(-1)min(-1) until death ensued. The EEG signal was acquired continuously; serial arterial plasma and terminal tissue thiopentone concentrations were measured enantiospecifically. Relevant drug tissue : plasma distribution coefficients and plasma concentration-EEG effect relationships were determined. 3. Doses (mg kg(-1)) (mean+/-s.e.mean) for anaesthesia (toe pinch) and lethality (respiratory failure), respectively, decreased in the order R-thiopentone (55.8+/-2.4 and 176.2+/-11.2)> rac-thiopentone (39.3+/-2.1 and 97.5+/-3.9)> S-thiopentone (35.6+/-1.9 and 74.2+/-5.2); plasma drug concentrations (microg ml(-1)) decreased in the order R-thiopentone (66.3+/-4.5 and 89.8+/-5.2)> rac-thiopentone (56.7+/-2.0 and 77. 8+/-2.8)> S-thiopentone (55.0+/-1.9 and 64.1+/-2.8). 4. Initial EEG activation was similar for all thiopentone forms. Plasma drug concentrations for the same extent of EEG deactivation reflected the potency order. 5. After infusion of rac-thiopentone, tissue : plasma distribution coefficients were higher for R- than for S-thiopentone in brain and visceral regions, but not in fat or muscle. After infusion of the separate enantiomers, the relative heart : brain distribution ratio was for S-thiopentone was double that for R-thiopentone. 6. The therapeutic index of R-thiopentone (3.16+/-0. 14) was more advantageous than either rac-thiopentone (2.52+/-0.13) or S-thiopentone (2.10+/-0.14), possibly due to the relatively greater distribution into CNS tissues than heart. The data suggest that R-thiopentone could make a satisfactory single enantiomer substitute for rac-thiopentone.