The validity of a radioimmunoassay kit for direct measurement of cortisol in the plaice is described. This technique was used to determine the cortisol concentration in adult male and female plaice in 100 microliter of either plasma or serum. There was no significant difference in the serum cortisol values determined by radioimmunoassay either with or without chromatography. Storage of whole blood at 4 degrees for varying periods up to 24 hr, before removal of the serum, did not affect the cortisol concentration. Decreases also did not occur in serum samples kept at room temperature for up to 5 days, although losses of 1-12% occurred after 7 days and 11-39% after 8 days. The daily variations in serum cortisol levels were examined in fish exposed at 11 degrees to 12 hr light: 12 hr dark over a 32-hr period and the existence of a 24-hr cortisol rhythm was observed. Blood samples were taken from the same six fish every 4 hr, each fish being bled nine times during the experiment. Blood samples were also taken every 4 hr from groups of seven fish, each fish during this experiment being bled only once. Measurement of monthly serum samples throughout 1 year showed cortisol concentrations at a maximum in April, during the peak spawning period. There was no significant difference between cortisol levels in male and female plaice, except in January, and no difference between serum and plasma values in either sex.