A central role for insulin-like growth factor-I (IGF-I) in the regulation of somatic growth is no longer a subject of debate. 1, 2 The jury, however, has yet to make a decision regarding the relative contributions of locally expressed IGF-I compared to that derived from the circulation. As investigators, we believe that we can definitively answer the questions we pose. The reality, however, is that despite our best efforts, our experiments often do not allow definitive answers. All too frequently several alternative scenarios are available to explain results. As with the legal system, therefore, we must rely on the “weight of the evidence”. Here we attempt to provide a balanced summary of the cases for and against local and endocrine IGF-I actions. The evidence arguing for local IGF-I actions is strong, and in our view, impossible to dismiss. Because IGF-I is a secreted protein that is carried in the circulation by IGF binding proteins (IGFBPs), there is no a priori reason to exclude IGF-I endocrine actions. Furthermore, the case for endocrine IGF-I actions is compelling. Historically, the generally accepted view of the nature of IGF-I action has vacillated like the hemlines of woman’s dresses and the width of men’s ties. In the first 25 years after the 1957 report by Salmon and Daughaday3 indicating that the growth-promoting actions of growth hormone (GH) were indirect, the substance (s) postulated to mediate GH action were assumed to act as hormones: they were thought to be synthesized at a single site, secreted into the circulation, and acted at distant sites. 4, 5 This activity was initially termed “sulfation factor”. By consensus in 1972 GH-dependent peptides with growth promoting activity were called somatomedins. However, with the recognition that somatomedin-C, the highly GH-dependent somatomedin purified from blood, had significant homology with insulin, IGF-I became the most widely used name (IGF-II has homology with IGF-I, but it is not highly regulated by GH). The assumption that somatomedin-C/IGF-I exerted endocrine actions was well-founded. It was identified and purified from blood, and the best data at the time indicated that the liver was the major, if not sole, site of synthesis. The fact that virtually all investigators studying GH mechanisms of action were endocrinologists undoubtedly also influenced the establishment of this concept. In the early 1980’s we observed that explants from multiple fetal mouse tissues released immunoreactive IGF-I into the culture media6 and that a large portion of this immunoreactive material was recognized by cell surface receptors which appeared specific for IGF-I. Later we showed that multiple adult rat tissues had IGF-I concentrations much higher than could be explained by the contribution from the blood and that IGF-I levels in many tissues were regulated by GH. 7 We postulated, therefore, that actions of IGF-I were predominately local, being exerted either on the cells of origin (autocrine actions) or on nearby cells (paracrine actions). With the isolation of the IGF-I cDNA, 8 studies using Northern analyses (eg, ref. 9) and in situ hybridization histochemistry (eg, ref. 10) indisputably showed that IGF-I expression was widespread in rodent and human tissues and, in significant part, dependent on GH. This ubiquitous IGF-I expression, accompanied by similarly widespread expression of type 1 IGF receptors (IGF1R; the main IGF-I receptor) and IGFBPs, made cellular signaling and regulation of IGF-I availability