The present study was undertaken to determine whether T3, estrogen, and 11-ketotestosterone could alter a specific population of GnRH-containing neurons, as indicated by a change in messenger RNA (mRNA) levels in sexually immature male tilapia, Oreochromis niloticus. Two weeks after castration, fish were assigned to four treatment groups. One group served as the control (sesame oil); a single ip injection of (T3; 5 μg/g), estradiol benzoate (EB; 5 μg/g), or 11-ketotestosterone (KT; 5 μg/g) was administered to the remaining three groups. Twenty-four hours after the injection, brains were collected and processed for in situ hybridization histochemistry using 35S-labeled 30-mer antisense oligonucleotide probes complementary to the GnRH-coding region of chicken II, salmon, and seabream GnRH. Computerized image analysis was performed to quantify mRNA concentrations, neuronal numbers, and neuronal size of the terminal nerve-nucleus olfactoretinalis, preoptic, and midbrain GnRH neurons. KT had no effect on any of the above neuronal parameters examined for salmon or seabream GnRH. Neither T3, EB, nor KT was effective to induce changes in midbrain chicken GnRH II mRNA concentrations, neuronal numbers, and neuronal size, indicating that an as yet unknown regulatory mechanism may operate midbrain GnRH neurons. T3 specifically suppressed the concentration of terminal nerve salmon GnRH mRNA, and EB significantly increased preoptic seabream GnRH neuronal numbers. These results are consistent with the hypothesis that thyroid hormone, by suppressing terminal nerve GnRH expression, promotes inhibition of sexual maturation. Furthermore, the failure of KT, a nonaromatizable androgen, to influence preoptic GnRH neurons emphasizes that an estrogenic pathway, at the onset of sexual maturation, is responsible for the recruitment of additional preoptic GnRH neurons that are fundamental to reproduction and behavior.