The literature catalytic-spectrophotometric ic method of selenium (Se) determination based on the catalytic action of Se(IV) on the sulfide reduction of methylene blue has been critically analyzed by means of ''fixed absorbance, fixed time and decolorization (induction) period'' methods. The logarithm of the induction period (time required to nullify A(660) Of MB, ti) was linearly correlated to Se concentration (C-Se) and this method yielded the best quantitative estimate of Se between 5.1-10.1 mu M. The hydrophobic methylene blue and the anionic sulfide both get seperated into a Cetyltrimethyl ammonium bromide (CTAB) cationic micelle, producing micellar catalysis. Selenosulfide reduction of methylene blue was also accelerated by CTAB near micellar concentrations, shifting the linear range of Se determination (Ln t(i) vs. C-Se) down to 0.6-6.4 mu M I Se and increasing the analytical sensitivity. The micellar catalysis, i.e., increased reaction rate, was due to the concentration effect of polysulfide or selenosulfide on the micellar surface. The developed procedure was applied to both synthetic solutions containing possible interfering ions and to an anodic slime of electrolytic copper refining plant yielding reproducible Se results of relative standard deviation <4%. The precise Se analysis of a standard fly ash sample (CRM-1633) containing (9.4+/-0.5) mu g Se g(-1) was feasible using the developed method. Both Se(TV) and Se(VI) can be analyzed, the latter after reduction.