This study was conducted to determine whether an optimum redox electrode potential existed for 24-hr hypothermic perfusion of rabbit kidneys. The perfusate consisted of a Ringer's-albumin solution to which was added varying amounts of the reducing agents, glutathione and ascorbate, either individually or in equimolar amounts. Electrode potential was monitored with a vitreous carbon electrode in relation to a silver-silver chloride reference cell, and kidney function was measured after preservation by connection to the circulation of a perfusor animal via a shunt. The best results were obtained using equimolar amounts of the reducing agents. Under these circumstances a definite optimum range for perfusate electrode potential was identified (Es = 40-70 mV) within which renal function was indistinguishable from unpreserved controls. Higher and lower perfusate electrode potentials were associated with significantly lower creatinine clearances. However, the explanation for these results appeared to be more complex than redox control alone, since kidney function was dependent not only on the redox potential of the perfusate but also on the reducing agents with which the adjustment had been made. Ascorbate proved to be significantly better than glutathione within the optimum potential range.