A pot experiment and afield trial were conducted to study the remediation of an aged field soil contaminated with cadmium, copper and polychlorinated biphenyls (PCBs) (7.67 +/- 0.51 mg kg(-1) Cd, 369 +/- 1 mg kg(-1) Cu in pot experiment; 8.46 +/- 0.31 mg kg(-1) Cd, 468 +/- 7 mg kg(-1) Cu, 323 +/- 12 microg kg(-1) PCBs for field experiment) under different cropping patterns. In the pot experiment Sedum plumbizincicola showed pronounced Cd phytoextraction. After two periods (14 months) of cropping the Cd removal rates in these two treatments were 52.2 +/- 12.0 and 56.1 +/- 9.1%, respectively. Total soil PCBs in unplanted control pots decreased from 323 +/- 11 to 49.3 +/- 6.6 microg kg(-1), but with no significant difference between treatments. The field microcosm experiment intercropping of three plant species reduced the yield of S. plumbizincicola, with a consequent decrease in soil Cd removal. S. plumbizincicola intercropped with E. splendens had the highest shoot Cd uptake (18.5 +/- 1.8 mg pot(-1)) after 6 months planting followed by intercropping with M. sativa (15.9 +/- 1.9 mg pot(-1)). Liming with S. plumbizincicola intercropped with M. sativa significantly promoted soil PCB degradation by 25.2%. Thus, adjustment of soil pH to 5.56 combined with intercropping with S. plumbizincicola and M. sativagave high removal rates of Cd, Cu, and PCBs.