Sex reversal, a mismatch between phenotypic and genetic sex, can be induced by chemical and thermal insults in ectotherms. Therefore, climate change and environmental pollution may increase sex-reversal frequency in wild populations, with wide-ranging implications for sex ratios, population dynamics, and the evolution of sex determination. We propose that reconsidering the half-century old theory "Witschi's rule" should facilitate understanding the differences between species in sex-reversal propensity and thereby predicting their vulnerability to anthropogenic environmental change. The idea is that sex reversal should be asymmetrical: more likely to occur in the homogametic sex, because sex-reversed heterogametic individuals would produce new genotypes with reduced fitness. A review of the existing evidence shows that while sex reversal can be induced in both homogametic and heterogametic individuals, the latter seem to require stronger stimuli in several cases. We provide guidelines for future studies on sex reversal to facilitate data comparability and reliability.

Populations of ectothermic vertebrates are vulnerable to environmental pollution and climate change because certain chemicals and high temperature can cause sex reversal during their larval development (i.e. genetically female individuals develop male phenotype or vice versa), which may distort population sex ratios. However, we have troublingly little information on sex reversals in natural populations, due to unavailability of genetic sex markers. Here we developed a genetic sexing method based on sex-linked single nucleotide polymorphism loci to study the prevalence and fitness consequences of sex reversal in agile frogs (Rana dalmatina). Out of 125 juveniles raised in laboratory without exposure to sex-reversing stimuli, 6 showed male phenotype but female genotype according to our markers. These individuals exhibited several signs of poor physiological condition, suggesting stress-induced sex reversal and inferior fitness prospects. Among 162 adults from 11 wild populations in North-Central Hungary, 20% of phenotypic males had female genotype according to our markers. These individuals occurred more frequently in areas of anthropogenic land use; this association was attributable to agriculture and less strongly to urban land use. Female-to-male sex-reversed adults had similar body mass as normal males. We recorded no events of male-to-female sex reversal either in the lab or in the wild. These results support recent suspicions that sex reversal is widespread in nature, and suggest that human-induced environmental changes may contribute to its pervasiveness. Furthermore, our findings indicate that sex-reversal is associated with stress and poor health in early life, but sex-reversed individuals surviving to adulthood may participate in breeding.