Mean daily temperature in natural nests of freshwater turtles with temperature-dependent sex determination is known to be a poor predictor of hatchling sex ratios when nest temperatures fluctuate. To account for this, a model was developed on the assumption that females will emerge from eggs when more than half of embryonic development occurs above the threshold temperature for sex determination rather than from eggs that spend more than half their time above the threshold. The model is consistent with previously published data and in particular explains the phenomenon whereby the mean temperature that best distinguishes between male and female nests decreases with increasing variability in nest temperature. The model, if verified by controlled experiments, has important implications for our understanding of temperature-dependent sex determination in natural nests. Both mean nest temperature and "hours spent above the threshold" will be poor predictors of hatchling sex ratios. Studies designed to investigate latitudinal trends and inter-specific differences in the threshold temperature will need to consider latitudinal and inter-specific variation in the magnitude of diel fluctuations in nest temperature, and variation in factors influencing the magnitude of those fluctuations, such as nest depth. Furthermore, any factor that modifies the relationship between developmental rate and temperature can be expected to influence hatchling sex ratios in natural nests, especially when nest temperatures are close to the threshold.
Keywords: Reptiles; Sex ratios; Temperature-Dependent sex determination; Thermal model.