The interrelationships between the corrosion rate of pipeline steel and soil parameters have been investigated under both controlled laboratory and natural outdoor conditions. Both defect-(holiday-) free, polyethylene-coated samples and samples with intentional one-square-centimeter holiday areas were studied. By using electrochemical impedance spectroscopy (EIS), the general behavior of the polyethylene coating could be distinguished from the general corrosion rate of the one-square-centimeter holiday. The laboratory samples were exposed for up to nine wet/dry cycles over a one-year period; whereas, the outdoor samples received only two major natural wettings during the one-year study. The defect-free, polyethylene-coated samples showed very little change during the one-year laboratory studies with two distinct soil types (one, a clay; the other, a sandy loam) and with moisture levels varying from near saturation (25 to 28% weight water), to dryness (less than 10% weight water). The corrosion rate of the exposed pipeline steel surfaces varied from ~ 2 muA/cm² (0.9 mpy) for the nearly water-saturated soil condition to ~ 2 x 10^-2</sup> muA/cm² (0. 009 mpy) for the dry soil case. Correlation between corrosion rates of laboratory and outdoor samples was good, and in general, the moisture level appears to control the corrosion rate in a semilog fashion from low water (typically 6% weight) to near saturation (typically 20 to 25% weight) concentrations. The corrosion rates in the two soil types were approximately equivalent at the same moisture levels. Combining these results with those of a previous study showed this correlation to be accurate for a broad range of soil types and conductivities. (Abstractor's note: relationships between soil type, corrosion and moisture concentrations are given. This may be useful for excavation purposes.)