The time course of GH secretion in response to hpGRF and its dependency on the extracellular Ca²⁺ concentration were studied in perifused dispersed anterior pituitary cells. The onset of GH secretion in response to 1nM hpGRF was relatively rapid (within 5s) but removal of hpGRF after 10-min application further increased the rate of secretion (off-response). The threshold and maximum concentrations of hpGRF in stimulatory secretion were 10^-12 and 10^-8M respectively. Between these two concentrations, the responses showed dose dependency. A reduction in the extracellular Ca²⁺ concentration to 0.25mM or to nominally zero reduced hpGRF-induced GH secretion to 64.4% or to 1.9%, respectively, of the control response in the presence of 2.5mM Ca²⁺. Two mM Co²⁺, known as a strong calcium channel blocker, completely suppressed hpGRF-induced GH secretion. The removal of Ca²⁺ from the perifusion buffer immediately after the offset of 1min-applied 1nm hpGRF accelerated the falling phase of GH secretion, which is parallel to the decline in [Ca²⁺]_o in the perifusion chamber. Under nominal Ca²⁺-free conditions, hpGRF produced no increase in GH secretion. However, 10min after the offset of 1min-applied hpGRF under Ca²⁺-free conditions, the introduction of normal buffer containing 2.5mM Ca²⁺ substantially restored GH secretion, although after 20min the introduction of normal buffer produced only a slight increase in GH secretion. In perifusion experiment of 10⁶ cells, intracellular cyclic AMP (cAMP) content was raised from the basal value of 4 to 26pmol by 2-min application of 1nM hpGRF. After cessation of hpGRF application, cAMP content decreased to 8.7pmol at 11min and returned to the basal value by 20min. The same tendency was observed in Ca²⁺-free buffer. In conclusion, the extracellular Ca²⁺ was essential for hpGRF-induced GH secretion. This indicates the importance of the influx of Ca²⁺ in response to hpGRF. The time course of hpGRF-induced rise and fall in cAMP content was roughly parallel to the GH secretion. The possible explanations of the off-response and the restoration of GH secretion by reintroducing normal buffer were discussed.