To establish normative data and determine the value of fluorometric AutoDELFIA assays (Wallac Oy) in the investigation of precocious puberty, we determined serum levels of LH, FSH, testosterone, and estradiol under basal and GnRH-stimulated conditions in 277 normal subjects at various pubertal stages and in 77 patients with precocious puberty. A substantial overlap was observed in basal and GnRH-stimulated gonadotropin levels in normal individuals of both sexes with pubertal Tanner stages 1 and 2. The 95th percentile of the normal prepubertal population was the cut-off limit between prepubertal and pubertal levels. These limits were 0.6 IU/L in both sexes for basal LH, 9.6 IU/L in boys and 6.9 IU/L in girls for peak LH after GnRH stimulation, 19 ng/dL in boys for basal testosterone, and 13.6 pg/mL in girls for basal estradiol. Basal and peak LH exceeding these limits were considered positive tests for the diagnosis of gonadotropin-dependent precocious puberty. According to these criteria, the sensitivities of basal and peak LH for the latter diagnosis were 71.4% and 100% in boys, and 62.7% and 92.2% in girls. The specificity and positive predicted value were 100% in both sexes for basal and peak LH levels. The negative predicted values for basal and peak LH were 62.5% and 100% in boys, and 40.6% and 76.5% in girls. Basal and GnRH-stimulated FSH levels overlapped among the various pubertal stages in normal subjects and were, in general, not helpful in the differential diagnosis of precocious puberty. In conclusion, basal LH levels were sufficient to establish the diagnosis of gonadotropin-dependent precocious puberty in 71.4% of boys and 62.7% of girls. In the remaining patients, a GnRH stimulation test was still necessary to confirm this diagnosis. Finally, suppressed LH and FSH levels after GnRH stimulation indicate gonadotropin-independent sexual steroid production.