This study was designed to investigate the central neuroendocrine mechanisms by which exercise (EX) stimulates growth hormone (GH) release as a function of age. Twelve male subjects, six in their early-to-mid twenties and six in their late sixties or seventies, received a strong GH stimulus either as incremental EX until volitional exhaustion or by administration of GHRH alone or Hex alone two hours after a presumed maximal GH response to combined administration of GHRH plus hexarelin (Hex). Total GH availability was calculated as area under the curve (AUC) over time periods 0 - 120 and 120 - 240 min. The mean AUC in µg/l × 120 min to GHRH+Hex in the younger group was approximately twice that in the older group (11,260, range 3,947 - 19,007 vs. 5,366, range 2,262 - 8,654). In younger males, the mean AUC to EX (509, range 0 - 1,151) was larger than to GHRH (119, range 0 - 543), but less than that to Hex (919, range 0 - 1,892). In the older group, GH responses to EX and GHRH were abolished (mean AUC: 112, range 0 - 285, and 156, range 30 - 493), respectively) in contrast to the response to Hex (1,077, range 189 - 1,780). These data indicate that maximal GH stimulation by GHRH+Hex results in greater desensitization of GHRH compared to Hex, irrespective of age. We postulate that the abolished responsiveness of GH to EX in older group is due to insufficient disinhibition of hypothalamic somatostatin activity and desensitization of GHRH, while the preserved activity of a central Hex-related pathway is not involved. The GH response to EX in younger males is due to complete inhibition of somatostatin activity and stimulation of a central Hex-related pathway in spite of GHRH desensitization. We conclude that a central Hex-related pathway is the primary factor for EX-induced GH release only in younger males.