In vitro working hearts of the turtle, Chrysemys picta bellii, paced at 30 beats/min, were studied over a range of input pressures in the following sequence of perfusion conditions: control normoxia, control anoxia, lactacidotic normoxia, and lactacidotic anoxia. Two such series of experiments were performed. In series 1 (n = 12), ventricular pressure (PV) and cardiac output were measured, and power output and dPV/dt were calculated. In series 2 (n = 5), intracellular phosphorus metabolites and intracellular pH (pHi) were also measured using 31P-nuclear magnetic resonance (31P-NMR) spectroscopy. In series 1 all mechanical variables increased with input pressure in generally similar fashion, except during anoxic acidosis, during which mechanical performance was depressed and was increased less or not at all by input pressure. Creatine phosphate (CP) and pHi fell significantly in anoxia and anoxic acidosis, but neither these variables, ATP, CP/ATP, nor, presumably, ADP changed as a function of input pressure with any perfusate despite often large increments in mechanical output. We conclude that anoxia and acidosis act synergistically to depress cardiac function in turtle hearts. Also, the insensitivity of NMR variables to changes in input pressure and cardiodynamics suggests that changes in these variables are unimportant for controlling energy turnover in this preparation.