The authors' objectives were as follows: 1) to characterize for the first time the relationship between whole body O2 delivery (DO2) and O2 consumption (VO2) in adult conscious dogs; and 2) to asses the effects of the inhalational anesthetic, halothane, on that relationship. DO2 was varied over a wide range in chronically instrumented dogs by gradual inflation and deflation of a hydraulic occluder implanted around the thoracic inferior vena cava to alter venous return and cardiac output. VO2 was measured at different values of DO2 in dogs in the fully conscious state and again during halothane anesthesia. A "binning" technique indicated that halothane decreased VO2 (P less than 0.01) at any given value of DO2 over a broad range of VO2. A two-line piecewise linear regression analysis technique indicated that halothane decreased (P less than 0.01) the critical O2 delivery (COD) from 20 +/- 3 to 10 +/- 1 ml.kg-1.min-1 and increased (P less than 0.01) O2 extraction at COD from 31 +/- 3 to 40 +/- 2%. However, the DO2-VO2 plots measured in both conscious and halothane-anesthetized dogs did not exhibit a discrete discontinuity but rather were closely fit (correlation coefficient = 0.98) by an exponential equation of the following form: O2 extraction = B1.(1 - exp (-DO2/B2))/DO2, where B1 is the delivery-independent estimate of VO2 and B2 is the "delivery constant," i.e., the DO2 associated with a VO2 equal to 63% of B1. Halothane decreased B1 (P less than 0.01) from 5.3 +/- 0.1 to 3.9 +/- 0.1 ml.kg-1.min-1 and decreased B2 (P less than 0.01) from 5.6 +/- 0.3 to 3.6 +/- 0.3 ml.kg-1.min-1 compared with that measured in conscious dogs. Thus, compared with the conscious state, halothane anesthesia alters the fundamental relationship between DO2 and VO2 and may have a beneficial effect on tissue oxygenation at low values of DO2.