The aim of this study was to establish the best method to measure copper (Cu) in serum using the atomic absorption spectrometry (AAS) apparatus (Rank Hilger, H 1550 type, Westwood Margate Kent CT9 4JL- England) with graphite tube. We compared the "standard curve" and the "extrapolation" methods and found for n=25, X±SD, SE as 78.2±0.07, 0.25 and 82.7±0.11, 0.35 pg/dL, respectively. There was not statistically any significant difference between them. In the end, because the standard curve was linear, recovery values and precision evaluations were better, the accuracy of the standard curve method was accepted to be good. In addition, it is a rapid and easily performed method, and is well suited for routine analysis of Cu in a clinical laboratory.