Groups of sexually undifferentiated sea bass Dicentrarchus labrax were fed with the androgen 17 -methyltestosterone (MT) during sex differentiation. MT treatment increased males from 79 3% in the controls (the usual 3 : 1 male : female sex ratio of cultured sea bass) to 100 0%, implying that in the treated groups one out of each five resulting males was a masculinized female (neomale). Thirteen males from the MT treated groups were taken as the parental generation and their sperm used to individually fertilize a pool of eggs from unrelated females. The probability of having at least one neomale was 95% and most probably two or three of the males used were neomales. The offspring from each family were reared separately under the same environmental conditions. Samples were taken at 11 and 15 months of age, during and after sex differentiation, respectively. Results showed that females predominated among the larger fish whereas males and undifferentiated fish predominated among the smaller ones. Intersexes exhibited an intermediate size. All fish with a body length smaller than 12 cm were undifferentiated. These results suggest that sex differentiation is more dependent on length than on age. At 15 months, sex ratios were male-biased in all families, except one (females ranged from 5 to 50%) and only two families had sex ratios not significantly different from 1 : 1, suggesting that the mechanism of sex determination in the sea bass is not of a XX/XY or ZW/ZZ type since no family exhibited a female-biased progeny, as would be expected from both types. Results support the hypothesis that factors other than genetic, i.e., environmental, may act epigenetically on the sex determination mechanisms of sea bass, as has been demonstrated in other fishes. 1999 The Fisheries Society of the British Isles