Nowadays the routine method for determination of nicotine in Iranian Tobacco Company is spectrometric method (ISO 2881: 1992). This study was performed in 2004 in order to compare some analytical methods. Three methods were conducted and compared, specifically methyl-t-butyl ether (MTBE) extraction (by GC); n-hexane extraction (by GC), and spectrometric method. A total of 20 samples of tobacco, flue-cured and Burley, with specified amount of nicotine (announced by their own company); were determined first by ISO 2881 method and then by two GC methods. After some experiments, it was shown that GC response for MTBE extraction is better than n-hexane extraction; so, the n-hexane method was omitted. In GC method tobaccos were extracted in alkaline media in presence of Internal Standard "quinoline". When the phases separate aliquots were taken and placed in autosampler vials for analysis. The GC response was calibrated using nicotine standard solutions. The percentage of nicotine was calculated from the peak area (or height), the graphs were plotted and the two series of results were compared by using analysis of variance or "t-test" method for P=0.05, and "correlation statistical calculation". The correlation factor (r) was 0.97 and t_calculated(1.98) was less than t_table (2.093). So the two methods have correlation with each other and there are no significant differences at the 95% confidence level.