Viviparity (live-bearing) is a reproductive mode that evolves from oviparity through gradual increases in the length of egg retention and thus intrauterine development. Squamate reptiles (lizards, snakes and amphisbaenians) provide a very useful model system for studying the evolutionary transition from oviparity to viviparity and the adaptive significance of this transition, because they exhibit a wide reproductive diversity. Approximately 20% of squamate reptiles are viviparous, and this reproductive mode has evolved far more often in this group of animals than in all other non-mammalian vertebrates combined. The selective forces responsible for evolution of viviparity remain unclear. To date, there are three main hypothesis of evolution of viviparity, cold-climate hypothesis, predictability hypothesis, and maternal manipulation hypothesis. Of these three main hypotheses, the maternal manipulation hypothesis is most widely accepted in scientific community, although empirical studies testing this hypothesis are still limited. Here, we used the Chinese skink Eumeces chinensis, a temperate-climate oviparous scincid lizard, as the model system to test the maternal manipulation hypothesis. Thirty-seven gravid females collected from a population in Lishui, Zhejiang province (eastern China) were maintained in the laboratory under three different thermal conditions until oviposition. Twenty-five females were provided with the basking opportunity, with 12 were allowed to bask for 14-h daily (LT treatment), and 13 for 10-h daily (ST treatment). The remaining 12 females were not provided with the basking opportunity (NTR treatment), and their body temperatures simply tracked changes in ambient environment. Females of different treatments did not differ morphologically. Body temperatures selected by females were measured using a UT325 digital thermometer. We measured each female three times, once on each of three consecutive days, beginng at 15:00 and ended within 2 h on each day. Females selected lower body temperatures and thermoregulated more precisely when gravid, and gravid females selected lower body temperatures than did non-gravid females and males. LT and ST females laid eggs earlier than did NTR females. Females maintained under the three thermal regimes did not differ from each other in any reproductive trait examined, including embryonic stage at oviposition. Six eggs randomly selected from each clutch were assigned to one of the following five treatments: in a room where temperatures varied naturally; in a constant-temperature room at 27 ℃; in two incubators where temperatures varied from 24-30 ℃ and 22-32 ℃, respectively; in the backyard of the laboratory mimicking natural nests. Incubation temperature (IT) affected incubation length, but maternal thermoregulation (MT) and the IT × MT interaction did not. MT affected the phenotype hatchlings, but IT and the IT × MT interaction did not. Hatchlings from eggs produced by LT females were longer in SVL, abdomen length and head length, and ran faster than those produced by ST females; hatchlings from eggs produced by NTR females had the shortest SVLs, abdomen lengths and head lengths, and performed worst in the racetrack. Other hatchling morphological traits examined did not differ among treatments. Our data support the two main predications from the maternal manipulation hypothesis: (1) gravid females shift their body temperature to provide the optimal thermal conditions for embryos; and (2) offspring phenotypes induced by maternal thermoregulation will enhance offspring fitness.