Developmental rates of western grapeleaf skeletonizer.Harrisina brilliansBarnes and MacDunnough. life stages were determined under constant temperatures ranging from 12.4 to 39°C. and linear and nonlinear developmental rate functions were fitted to the data.H. brilliansdevelopment under fluctuating temperatures also was investigated. with emphasis on high-temperature regimes. Degree-day and nonlinear methods for simulating larval development were validated by comparing predictedH. brilliansdevelopment with actual field observations. Assuming linear development, the lower development thresholds for egg, total larval, and pupal stages were 9.0, 8.8, and 9.2°C, respectively. Linear (degree-day) predictions of larval development in the field were more accurate than predictions derived by the nonlinear method. The linear model underestimated larval development at low temperatures and overestimated development at high temperatures by as much as 8.6%. The nonlinear model consistently underestimated larval development by as much as 17.3%. We suggest that error involved in nonlinear, temperature-dependent development models may relate to the critical assumption that enzyme systems respond almost instantaneously to temperature changes.