Several methods were used to study survival of Salmonella typhimurium LT2 in soil. An ion exchange resin-based extraction method was used to concentrate biomass from soil, from which DNA was extracted in order to quantify a Salmonella-specific sequence by a quantitative polymerase chain reaction (QPCR). S. typhimurium LT2 was detected at a minimum density of 10 Q cells g 3I in non-sterile soil, and the method proved to be specific for this organism in microcosm experiments. Non-sterile soil microcosms were inoculated with S. typhimurium LT2 at 10 U cfu g 3I dry soil and survival monitored at three matric potentials. Viable counts on XLD indicated a rapid decline in cell density over 54 days, whereas direct counts of active cells using the respiration-sensitive dye 5-cyano-2,3-ditolyl tetrazolium chloride (CTC) remained relatively constant. XLD did not underestimate culturable cells in comparison to non-selective agar. QPCR revealed that the number of salmonella targets (H-li) remained constant up to day 13. After that time, a decrease occurred, corresponding to that of the plate counts, due to an increase in resistance of the cells to lysis, as incorporation of a lysozyme step into the DNA extraction method allowed more efficient DNA extraction. This resulted in a constant QPCR signal over 54 days which correlated with direct, active cell counts. QPCR showed H-li was present at levels only slightly lower than those at day 0. There was no difference in survival between the three different moisture regimes. Direct CTC counts of S. typhimurium LT2 in the soil microcosms confirmed that intact cells were present in a metabolising state after 54 days in non-sterile soil, indicating a significant proportion of uncultured but active cells. z