Microwave energy, generated in a commercial microwave system designed for extraction of organic samples, has been used for the extraction of polycyclic aromatic hydrocarbons (PAHs) from highly contaminated land samples. Initial studies used a native soil and compared a 20 min microwave extraction with a 6 h conventional Soxhlet extraction (both using dichloromethane) with comparable results. The low concentration of PAHs found in the soil resulted in large relative standard deviation (s_r) values for the individual PAHs (maximum of 60.9% for microwave extraction). A second soil, with higher PAH levels, was microwave extracted using both dichloromethane and different compositions of an acetone–hexane mixture and compared with Soxhlet extraction. The results indicate that microwave extraction using pure acetone is a more efficient procedure than other combinations studied. A repeatability study gave an s_r value for the total PAHs extracted of 2.4%(n= 6). A central composite experimental design was used to elucidate optimum operating parameters but variations in temperature, extraction time and solvent volume were found to have little effect on recovery. The over-all s_r for the design was 5.34% illustrating no statistical significance in recovery. A third, Laboratory of the Government Chemist, test soil was used which indicated that the soil matrix was the greatest factor affecting PAH recovery.