Mussel species Mytilus galloprovincialis exhibit considerable gonad color variation ranging from light creamy white to dark orange. The coloration of gonadal tissue was previously thought to be a reliable indicator of sex allowing a visual determination of gender. Although Mediterranean mussel is a highly tolerant species, extreme environmental conditions cause oxidative stress and affect the accumulation of carotenoids in the gonads. Carotenoid pigments, liable for the orange coloration of gonads, are known for critical rolls in defense against oxidative stress, as they can convert harmful oxygen radicals into less-damaging hydrogen peroxide. Males can also display high concentrations of these pigments according to their position in the mussel bed, bearing in mind that visual identification of sex could lead to misinterpretation of results. Research has been performed to determine the reliability of visual identification of sex by comparing the results with the results obtained by the histological identification of sex as the most reliable method. Both sexing methods were applied to 40 mussels within two sites: a polluted rocky shore in Kaštela Bay (Croatia) under tidal stress and a mussel farm in the Bay of Strunjan (Slovenia), permanently immersed in the sea. Due to sex determination methods, two experimenters were assigned to establish the sex of each mussel twice. Visually, sex was determined according to the color of the mantle. Histologically, sex was determined according to the presence of sperm or ova and was used to quantify the percentage of mussels that would have been sexed incorrectly based on color observation. Histology showed that in polluted rocky shore, 42.5% were females and 57.5% were males, while in mussel farm 37.5% were females and 62.5% were males. Based on the observation of gonadal coloration, the frequency of females in both sites was higher which led to misinterpretation of results. In polluted rocky shore 52.5% observed mussels were females, while in mussel farm this percentage was 65.0%. Farm mussels were inaccurately sexed based on color more frequently (52.5%) than rocky shore mussels (40.0%). In both sites, of all misinterpreted mussels, more males than females were incorrectly sexed. Farm males were sexed incorrectly 76.2% of the time, which was more often than rocky shore males (62.5%). Most of misinterpreted males in both sites had orange gonad tissue, which could be related to an increased concentration of carotenoid pigments. According to the obtained results, the visual method of sex determination is not a reliable, therefore it is necessary to perform histological tests in such studies.