Salmonella cells in two sugar-rich media were heat treated at various constant temperatures in the range of 55 to 80 degrees C and their survival ratios determined at various time intervals. The resulting nonlinear semilogarithmic survival curves are described by the model log10S(t) = -b(T)tn(T), where S(t) is the momentary survival ratio N(t)/N0, and b(T) and n(T) are coefficients whose temperature dependence is described by two empirical mathematical models. When the temperature profile, T(t), of a nonisothermal heat treatment can also be expressed algebraically, b(T) and n(T) can be transformed into a function of time, i.e., b[T(t)] and n[T(t)]. If the momentary inactivation rate primarily depends on the momentary temperature and survival ratio, then the survival curve under nonisothermal conditions can be constructed by solving a differential equation, previously suggested by Peleg and Penchina, whose coefficients are expressions that contain the corresponding b[T(t)] and n[T(t)] terms. The applicability of the model and its underlying assumptions was tested with a series of eight experiments in which the Salmonella cells, in the same media, were heated at various rates to selected temperatures in the range of 65 to 80 degres C and then cooled. In all the experiments, there was an agreement between the predicted and observed survival curves. This suggests that, at least in the case of Salmonella in the tested media, survival during nonisothermal inactivation can be estimated without assuming any mortality kinetics.