Our understanding of the testicular control of growth and functioning of the accessory sex glands began with an observation in the 18th century of John Hunter (1), who discovered in animals the endocrine dependency of the prostate. He demonstrated that castration in experimental animals causes a decrease in the volume of the prostate. White (2) reported in 1895 favourable results of "double" castration in men with hypertrophy of the prostate and one year later Cabot (3) proposed castration as treatment for an enlarged prostate in general. Differentiation between carcinoma and BPH was not done. The studies of Huggins and Hodges (4,5) in the early 1940's defined the regulatory role of the testes and testicular androgens in prostatic cancer. Their work placed the orchiectomy for treatment of prostatic cancer on a scientific basis and brought its general acceptation. Since then different forms of hormonal therapy have been used in all stages of prostatic cancer. Today, in the 80's, endocrine manipulation is generally accepted as first treatment in disseminated disease, although the impact of hormonal manipulation on patient survival (6,7) and the optimal timing of the initation of treatment are still controversial (8-10). After more than 45 years of investigation no single hormonal treatment has proven to be superior, so the conclusion of Scott in an overview (11) on hormonal therapy for prostatic cancer: "We have gone as far as we can go in the hormonal treatment of advanced prostatic cancer, and it is unlikely that further search will reveal a better treatment than castration- estrogen therapy", may still be valid. This statement, although seemingly definite, bears the challenge to develop modalities of treatment which are more effective than endocrine manipulations. As a part of future treatment it will be necessary to find methods that enable us to predict which patient will and which patient will not have a prolonged response to hormonal treatment. The estimation of nuclear androgen receptor (ARn) seemed promissing according to the findings of Trachtenberg and Walsh (12) and Ghanadian et al (13). The present thesis is aimed to contribute to this problem.