There is so far no evidence that the newest hormone compounds, such as progestational agents, antiandrogens or estramustine phosphate, yield better results than low dose estrogens or orchidectomy in previously untreated patients. Castration and the other modalities of androgen deprivation, although they can cause a measurable objective regression in less than half of cases, do often induce a marked, sometimes long-lasting subjective improvement and can delay progression. It has not yet been clearly established whether or not hormone therapy can bring about a prolongation in survival. The critical point still remains our poor understanding of the causes of acquired hormone resistance and our inability to prevent it. Further studies on hormone receptors and steroid metabolism in normal and neoplastic cells might lead to progress in this field. The unsolved problem, whether resistance is due to a selection of clones resistant to hormones ab initio or to acquired biochemical characteristics in cells that were initially responsive, might be of great practical significance. A reliable test, able to detect sensitivity or resistance towards hormone therapy before any treatment is given, would be extremely helpful. Unfortunately, neither the most sophisticated hormone balance studies nor receptor determinations have led to clinically useful implications.