A field experiment was conducted in Delaware (USA) to evaluate three crude oil bioremediation techniques. Four treatments were studied: no oil control, oil alone, oil + nutrients, and oil + nutrients + an indigenous inoculum. The microbial populations were monitored by standard MPN techniques, PLFA profile analysis, and 16S rDNA DGGE analysis for species definition. Viable MPN estimates showed high but steadily declining microbial numbers and no significant differences among treatments during the 14-weeks. Regarding the PLFA results, the communities shifted over the 14-week period from being composed primarily of eukaryotes to Gram-negative bacteria. The Gram-negative communities shifted from the exponential to the stationary phase of growth after week 0. All Gram-negative communities showed evidence of environmental stress. The 16S rDNA DGGE profile of all plots revealed eight prominent bands at time zero. The untreated control plots revealed a simple, dynamic dominant population structure throughout the experiment. The original banding pattern disappeared rapidly in all oiled plots, indicating that the dominant species diversity changed and increased substantially over 14 weeks. The nature of this change was altered by nutrient-addition and the addition of the indigenous inoculum.