Copper (Cu) and Zinc (Zn) in serum of 113 healthy controls (43 males and 70 females) and of 68 patients with rheumatoid arthritis (19 males and 49 females) were determined using a HITACHI MODEL 207 atomic absorption spectrophotometer. Wavelength and lamps current of Cu and Zn were setted at 3247 A, 2139 A, and 10mA, 9mA respectively. Flow rate of acetylene and air were setted at 3.0 l/min, and 13.0 l/min. The stock standard solutions of Cu and Zn were 0.5 mg, 1.0mg, 2.0mg and 3.0mg per 100mE in 0.01 N·HCl. For working standard solutions, these stock standard solutions were diluted to 50μg/100ml. 100μg/100ml, 200μg/100ml and 300μg/100ml with 20% glycerine. Serum and working standard so· lutions of Cu and Zn were diluted five fold with deionized distillation water, then aspirated directly into the burner. The relative absorbance was expressed as scale reading by a HITACHI MODEL QPD-54 recorder. Various acids and other elements had few influences on the determi· nations of Cu and Zn. The mean recovery rate of Cu and Zn in serum were 100.5% and 100.8% respectively. By this method, the levels of Cu and Zn in healthy controls were 103.0±14.5 (S.D.), 116.2±18.6 (S.D.) μg/100ml in males and 108.7±19.6 (S. D.), 110.3±14.7 (S.D.) μg/100ml in females respectively. No sex difference were observed. The ratio of Cu and Zn (Cu/Zn ratio) of healthy controls were 0.89±0.17 (S. D.) in males and O.99±0.25 in (S.D.) females. The sex differrences of Cu/Zn ratio were statistically significant (p<0.05). The levels of Cu and Zn in rheumatoid arthritis were 141.7±25.2 (S.D.). 89.1±14.5 (S.D.) /-Lg/100 mP and 154.l±29.7 (S.D.), 86.7±18.1 (S.D.) μg/100ml in females respectively. No sex differren· ces were observed. The Cu/Zn ratio were 1.64±0.43 (S.D.) in males and 1.82±0.48 (S.D.) in females, wich showed no sex differrences. In patients with rheumatoid arthritis, serum Cu levels were significantly higher (P<0.001) and serum Zn levels were significantly lower (P<0.001) than in healthy controls in both sexes. The Cu/Zn ratio were significantly higher (P<0.001) rheumatoid arthritis than in controls. The serum Cu in 13 of 19 males (68%), and 30 of 49 females (61 %), with rheumatoid arthritis were shown to be above the upper limit of 5% rejection limit in healthy controls. The serum Zn in 3 of 19 males (16%), and in 24 of 49 (47%) with rheumatoid arthritis were shown to be below the lower limit 5% rejection limit in healthy controIs. The Cu/Zn ratio in rheumatoid arthritis, in 14 of 19 males (74%), and in 35 of 49 females (71%) were shown to be above the upper limit of 5% rejection limit in healthy of males and females. So, differences in Cu/Zn ratio between healthy controls and rheumatoid arthritis were proved to be more pronounced.