A study of sera from 285 patients with definite or classical rheumatoid arthritis (including 37 patients receiving no anti-inflammatory drugs) and sera from 67 healthy subjects has confirmed 10 published reports of a statistically significant decreased blood histidine concentration in patients with rheumatoid arthritis. Contrastingly, in sera from 231 patients with a variety of acute and chronic illnesses other than rheumatoid arthritis, no statistically significant hypohistidinemia was observed either in the group as a whole or in association with the administration of aspirin, prednisone, indomethacin, phenylbutazone, or dextropropoxyphene. In the patients with rheumatoid arthritis there was a statistically significant correlation between the serum histidine concentration and the following: Westergren sedimentation rate (r=-0.33, P smaller than 10- minus 9), grip strength (r=0.26, P smaller than 10- minus 9), hematocrit (r=0.23, P smaller than 10- minus 9), duration of morning stiffness (r=-0.14, P=10- minus 5), walking time (r=-0.13, P=10- minus 4), latex titer of rheumatoid factor (r=-0.11, P=0.001), and the duration of arthritis (r=-0.06, P=0.05). There was no statistically significant association between the serum histidine concentration and the duration of rheumatoid arthritis in the 151 patients with disease of 0-10-yr duration (r=0.02, P=0.5), the sex of the patient, or the presence of antinuclear antibody (R=0.007, P=0.9). The serum histidine concentration was less in rheumatoid patients receiving steroids (P=0.00001), gold (P=0.009), and aspirin (P=0.15) than in rheumatoid patients not receiving these drugs. This study indicates that histidine determinations on properly preserved casual serum samples can be helpful in the diagnosis of rheumatoid arthritis and in the evaluation of the activity of the disease.