In order to simply radioimmunoassay for plasma testosterone and to measure many samples at the same time, a method of solid phase radioimmunoassay utilizing a plastic disposable microtiter tray (DMT) by which chromatography can be omitted was investigated. The antiserum was obtained by immunizing rabbits with testosterone-3 BSA which had been synthesized according to the Erlarnger's method. Plasma samples (male: 0.05ml, female: 0.2 ml) were extracted with 1.0 ml of ether. After freezing the plasma layer in an acetone-dryice bath, the ether phase was transfered to a glass tube and evaporated to dryness. These samples and the dried standard testosterone were dissolved with borate buffer containing 3H-testosterone and transfered to plastic DMT which had been precoated with the diluted antiserum, and incubated for 24 hrs. After removal of the incubated solution, the cups of DMT were cut off and were dissolved with toluene scintillator in counting vials. The radioactivity was counted with a liquid scintillation counter. Other steroids except for 5alpha-dihydrotestosterone (5alpha-DHT) had a low degree of cross reactivity with the antiserum. Five alpha-DHT which could be measured together with testosterone in this assay was not a problem clinically because of its strong androgenic activity. The best standard curve was obtained when the antiserum was diluted to 1:1000. The sensitivity of this assay was 10 pg/tube. The maximal adsorption of antibody to plastic DMT was observed when the pH of antiserum was within the range of 6.5-9.5 and the precoating time was 24 hr at room termperature. The best pH of incubation buffer was 8.0, and the antigen-antibody reaction became a plateau when the incubation exceeded 6 hrs. Water blank in this assay was 4.6 +/- 2.1 pg/tube. The recovery of testosterone (50, 100, 200 pg) added to 0.1 ml female plasma was 99 +/- 6.8%. Coefficients of variation within assay and between assay were below 11.2% and 20.0%, respectively. Correlation between this method and the dextran-coated charcoal method was fairly good (r=0.938). Plasma testosterone levels in 10 normal males and 12 normal females were 616 +/- 202 (mean +/- SD) ng/dl and 66 +/- 29 (mean +/- SD) ng/dl, respectively. The levels were low in patients with hypopituitarism, hypogonadism and acromegaly. They were normal in patients with Cushing's syndrome due to adrenal hyperplasia and adenoma, but they were high in a patient with adrenal carcinoma. In a patient with testicular feminization, the level was 632 ng/dl. This increased after the administration of HCG, and decreased to 127.5 ng/dl after castration. This solid phase radioimmunoassay (using plastic DMT) is economically feasible as well as simple because it is possible to separate the bound hormone from the free hormone of all samples at the same time and there is little restriction in time and temperature. According to the above results, this method is suitable for routine clinical use.