Ghrelin, an acylated peptide produced predominantly by the stomach, has been discovered to be a natural ligand of the growth hormone secretagogue receptor type 1a (GHS-R1a). Ghrelin has recently attracted considerable interest as a new orexigenic factor. However, ghrelin exerts several other neuroendocrine, metabolic and also nonendocrine actions that are explained by the widespread distribution of ghrelin and GHS-R expression. The likely existence of GHS-R subtypes and evidence that the neuroendocrine actions, but not all the other actions, of ghrelin depend on its acylation in serine-3 revealed a system whose complexity had not been completely explored by studying synthetic GHS. Ghrelin secretion is mainly regulated by metabolic signals and, in turn, the modulatory action of ghrelin on the control of food intake and energy metabolism seems to be among its most important biological actions. However, according to a recent study, ghrelin-null mice are neither anorectics nor dwarfs and this evidence clearly depicts a remarkable difference from leptin null mice. Nevertheless, the original and fascinating story of ghrelin, as well as its potential pathophysiological implications in endocrinology and internal medicine, is not definitively cancelled by these data as GHS-R1a null aged mice show significant alterations in body composition and growth, in glucose metabolism, cardiac function and contextual memory. Besides potential clinical implications for natural or synthetic ghrelin analogues acting as agonists or antagonists, there are several open questions awaiting an answer. How many ghrelin receptor subtypes exist? Is ghrelin 'the' or just 'a' GHS-R ligand? That is, are there other natural GHS-R ligands? Is there a functional balance between acylated and unacylated ghrelin forms, potentially with different actions? Within the next few years suitable answers to these questions will probably be found, making it possible to gain a better knowledge of ghrelin's potential clinical perspectives.