In bovine retinal rods, transducin loaded with GTP or GTP gamma S (T*) activates a cGMP phosphodiesterase (PDE) by forming a tightly membrane-bound complex with it [Catty, P., et al. (1992) J. Biol. Chem. 267, 19489-19493]. Up to two T*s are able to bind to PDE [Clerc, A., & Bennett, N. (1992) J. Biol. Chem. 267, 6620-6627]. We analyze here PDE activation by two successive bindings of T*. In the mathematical model used, we took into account that the membrane concentration determines the amount of PDE able to interact efficiently with T* through the attachment of PDE itself to the membrane. We therefore fitted the data obtained over a wide range of membrane and PDE concentrations. We found that the binding of the first T* to PDE elicits 80-100% of the maximal activity of PDE, whereas the binding of the second T* to PDE elicits little or no additional activation of PDE. This finding profoundly differs from previous conclusions. The carefully controlled conditions of our experiments permit one to understand these discrepancies. In the physiological situation, PDE would be nearly maximally activated through its interaction with only one T*. The efficient binding of the second T* to those complexes would then ensure a rapid deactivation of T* through the enhancement of the rate of GTP hydrolysis in T* bound to PDE [Pagès, F., et al. (1992) J. Biol. Chem. 267, 22018-22021; Pagès, F., et al. (1993) J. Biol. Chem. 268, 26358-26364].