Beta 3-adrenoceptor (beta 3-AR) agonists were found to have remarkable anti-obesity and anti-diabetic effects in rodents shortly after their discovery in the early 1980s. Despite these promising qualities, several pharmaceutical problems and theoretical concerns have slowed the development of these products as therapeutic agents in humans during the last 15 years. To date, the pharmaceutical industry has not been successful in developing a beta 3-AR agonist for use in the treatment of human obesity and type 2 diabetes. Pharmaceutical problems in this area concern important differences between rodent and human beta 3-AR and the difficulty in finding a compound with sufficient bioavailability that is a highly selective and full agonist at the human receptor. Some of these problems seem to have been solved with the cloning of the human beta 3-AR, which has made it possible to develop novel compounds directly and specifically against the human receptor. However, several theoretical concerns still remain. These include the major question as to whether the number of biologically active beta 3-ARs in adult humans is sufficient to produce relevant metabolic effects and, if so, whether their long-term stimulation is safe and free of unwarranted side effects. In addition, the mechanisms of action of beta 3-AR agonists remain poorly understood. Recent studies using CL 316,243, a highly selective beta 3-adrenergic compound, have provided new insights into the potential mechanisms of action of these drugs in rodents as well as the first evidence that treatment with a highly selective beta 3-AR agonist exerts relevant metabolic effects in humans. It appears that chronic beta 3-adrenergic stimulation in white adipose tissue increases the expression of newly discovered mitochondrial uncoupling proteins (UCP 2 and 3) and a "reawakening" of dormant brown adipocytes. In addition, beta 3-ARs may be present in skeletal muscle where ectopic expression of UCP-1 has been reported. If these findings are confirmed, tissues other than brown fat may play an important role in mediating beta 3-adrenergic effects on thermogenesis and substrate oxidation. In humans, treatment with CL 316,243 for 8 weeks, in spite of limited bioavailability, induced marked plasma concentration-dependent increases in insulin sensitivity, lipolysis, and fat oxidation in lean volunteers, without causing beta 1-, or beta 2-mediated side effects. These results clearly indicate that favourable metabolic effects can be achieved by selective beta 3-AR stimulation in humans. The compounds of the next generation currently emerging from preclinical development are full agonists at the human beta 3-AR. These agents have demonstrated promising results in non-human primates. It will be interesting to see whether their efficacy in clinical trials is superior to that achieved with previous (rodent) beta 3-AR agonists and, if so, whether their effects will eventually translate into weight loss and improved metabolic control that could facilitate their use as effective drugs for the treatment of obesity and Type 2 diabetes in humans.