Field-measured heritability of the threshold for sex determination in a turtle with temperature-dependent sex determination

Problem: For temperature-dependent sex determination to respond to selection, there should be genetic variance underlying the threshold that switches development from a male-producing program to a female-producing program. Genetic variance for this threshold in reptiles has never been estimated under field conditions. Methods: We estimated variance components of the thermal sensitivity of the sex determination threshold under field conditions for the painted turtle, Chrysemys picta, a species that has temperature-dependent sex determination. Multiple paternity within clutches was identified by genotyping females and their offspring. We endeavoured to statistically account for common nest and maternal effects and estimated the genetic variance underlying the sex determination threshold under field conditions by isolating the contribution of sires within a single clutch. Results: With 51 clutches containing 393 offspring, we estimated significant heritability for the sex determination threshold (h(2) = 0.351, 95% CI = [0.164, 0.8321). Using a more restrictive dataset, which included only those clutches where each sire was represented by at least two offspring (34 nests, 273 hatchlings), heritability was not significantly different from zero (h(2) = 0.173, 95% CI = [0.000, 0.628]). Paternal siring success did not influence hatchling sex; thus, we have no evidence to support differential sex allocation across different sires within the same nest. Conclusion: We used a natural 'breeding design' under field conditions to show that the threshold of temperature-dependent sex determination may have a heritable genetic basis.