Does the variance of incubation temperatures always constitute a significant selective force for origin of reptilian viviparity？

To test the hypothesis that the variance of incubation temperature may have constituted a significant selective force for reptilian viviparity, we incubated eggs of the slender forest skink Scincella modesta in five thermally different natural nests and at two constant temperatures （18℃ and 21 ℃）. Our manipulation of incubation temperature had significant effects on incubation length and several hatchling traits （snout-vent length, tail length, fore-limb length, and sprint speed）, but not on hatching success and other hatchling traits examined （body mass, head size, and hind-limb length）. Incubation length was nonlinearly sensitive to temperature, but it was not correlated with the thermal variance when holding the thermal mean constant. The 18 ℃ treatment not only produced smaller sized hatchlings but also resulted in decreased sprint speed. Eggs in the nest with the greatest proportion of temperatures higher than 28 ℃ also produced smaller sized hatchlings. None of the hatchling traits examined was affected by the thermal variance. Thermal fluctuations did result in longer incubation times, but females would benefit little from maintaining stable body temperatures or selecting thermally stable nests in terms of the reduced incubation length. Our data show that the mean rather than the variance of temperatures has a key role in influencing incubation length and hatchling phenotypes, and thus do not support the hypothesis tested .