Two response surface models, a third-order polynomial and a modified Gompertz function, were evaluated for their ability to describe germination and infection of Bremia lactucae on lettuce in relation to temperature and leaf wetness duration. The models were based on data obtained from three independent experiments in which lettuce plants were inoculated with the fungus and exposed to a range of leaf wetness durations (0-24 h) at each of six fixed temperatures (5-30-degrees-C). Germination was assessed at the end of each wetness period, and infection was evaluated by measuring disease severity 10 or 14 days after inoculation. In the Gompertz model, rate and asymptote parameters could be expressed as simple functions of temperature using quadratic and cubic functions, respectively, and parameter estimates were obtained in a simple three-step procedure. Models based on the modified Gompertz function had non-significant lack-of-fit statistics, and yielded higher correlations between observed and predicted germination and infection levels than third-order polynomials. The presented methodology may be adapted to model processes in other biological and ecological systems in which fast early growth or development is observed in relation to temperature (or a similar factor) and a time-like variable.