GH release is normally stimulated by the naturally occurring GH-releasing factor (GRF). However, smaller GH-releasing peptides (GHRPs) and non-peptide analogues have been described which stimulate GH release in animals and man. Although these compounds release GH in vitro, their in vivo activity in conscious animals has proved more difficult to study since the GH responses are variable, and prone to desensitization. We now compare the GH-releasing properties of GHRP-6 and a novel benzolactam GH secretagogue L-692,585 using chronically cannulated guinea pigs and automated blood micro-sampling to study the effects of repeated exposure to these secretagogues. L-692,585 was approximately tenfold less potent than GHRP-6 for GH release, but it synergized strongly with GRF. Serial injections of GRF, GHRP-6 or L-692,585 at intervals of 60 or 90 min produced variable GH release which followed a cyclic pattern of responsiveness. Prolonging the pulse interval to 3 h produced more regular responses to both GHRP-6 and L-692,585. Continuous i.v. infusion of low doses of either secretagogue elicited an initial GH release, and amplified the spontaneous GH secretory pattern over the next 6 h. We conclude that L-692,585 and GHRP-6 share similar in vivo, their in vivo activity in conscious animals has frequent injections is similar for all three secretagogues, and is a property of the conscious animal rather than of any secretagogue type. More consistent responses can be obtained with less frequent injections that more closely match the endogenous GH rhythm, whereas continuous exposure to these secretagogues leads to amplified endogenous secretion. Our results show that the interpretation of in vivo effects of these peptide and non-peptide secretagogues will need to take account of their interaction with the endogenous mechanisms governing GH release.