In this study, a new method was developed for the pre-concentration of trace mercury from seafood samples prior to analysis by spectrophotometry. The method is based on the complexation between 2-hydrazinobenzothiazole (2-HBT)-modified copolymer(s) and Hg(II) in the presence of an ionic surfactant, cetyltrimethylammonium bromide (CTAB), as sensitivity enhancer at pH4.5 and the extraction of the complex into the surfactant-rich phase of polyethylene glycol tert-octylphenyl ether (Triton X-114) as the extractant. The variables affecting extraction efficiency were evaluated and optimized. Due to the observation that the modified copolymers are 2.5-fold more sensitive and selective to the Hg2+ ions than the CH3Hg+, the amounts of free Hg2+ and total Hg were determined at 325nm by spectrophotometric detection of free Hg2+ and total Hg in the pre-treated and extracted fish samples using dilute acid mixture containing Triton X-114 and K2Cr2O7, before and after oxidation of CH3Hg+ to Hg2+ with mixture of KBr and KBrO3 in the acidic media. The amount of CH3Hg+ was calculated from the difference between total Hg and free Hg2+ amounts. The accuracy was tested by analysis of two certified samples. The results were statistically in good agreement with the certified values, and the precision was lower than 6.4%. The limits of detection were 1.40 (1.58) and 1.91 (2.11)mu gL(-1) for Hg2+ from the two calibration solutions spiked before the pre-treatment, respectively. It has been observed that there is no significant matrix effect by comparison of slopes of the calibration curves. The method was applied to seafood samples for speciation analysis of free Hg2+ and CH3Hg+. In terms of speciation, while total Hg is detected in the range of 12.6-143.8 mu gkg(-1), the distribution of mercury in seafood was in the range of 7.4-53.3 mu gkg(-1) for CH3Hg+ and in 8.3-90.5 mu gkg(-1) for free Hg2+.