The Charnov-Bull hypothesis has been widely accepted to explain the adaptive significance of temperature-dependent sex determination (TSD), but empirical support for this hypothesis is still limited. Eggs of the Chinese three-keeled pond turtle (Chinemys reevesii) were incubated at 26 ℃, 28℃, and 30℃, and treated with 17β-estradiol or Fadrozole (an aromatase inhibitor) to obtain both normal and sex-reversed hatchlings at each temperature. Our objectives were to test the Charnov-Bull hypothesis by assessing the influence of incubation temperature and sex on hatchling traits. Incubation duration was longer in males than females at 26℃, but did not differ between the sexes at 28℃ and 30℃. Hatchlings from low temperatures were larger than those from high temperatures, and females were larger than males. The swimming capacity of hatchlings was not affected by either incubation temperature or sex. Hatchling growth was not influenced by incubation temperature, but differed between the sexes. Females grew faster than males. The Charnov-Bull hypothesis predicts that in the male-female pattern of TSD, females from high temperatures should have higher fitness than those from low temperatures and/or that males from high temperatures should have lower fitness than those from low temperatures. Our results are inconsistent with this prediction and thus give no support to this hypothesis.