A suite of salt-marsh peat samples from four sites along the coast of Maine (Wells, Phippsburg, Gouldsboro, and Machiasport) has been analyzed using high-precision techniques to determine local relative sea-level trends and to evaluate proposed along-coast warping. A spatially variable set of relative sea-level records in Maine would have important implications for geophysical models that predict the response of the lithosphere during deglaciation and postglacial isostatic relaxation. These models are often at odds with observed relative sea-level indicators near the margins of former glaciation, including those from Maine. Assemblages of agglutinated benthic foraminifera occur in vertical zones on the surface of modern salt marshes in Maine and can be used to accurately locate former mean high water levels in cores. Additional tools in this study include accelerator mass spectrometer 14C dating of individual plant fragments and precise leveling of elevations. The amplification of M2 tidal range in the Gulf of Maine and the Bay of Fundy during the Holocene is modeled and applied to the mean high water data yielding best-estimate envelopes of mean tide level change for each location. Average long-term (thousands of years) mean tide level rise did not exceed ???2 mm/yr at any time during the late Holocene at Wells, Phippsburg, and Machiasport. Between 4.5 and 3 ka (calibrated [cal]), the apparent rate of rise at Gouldsboro was higher than at any other site studied. This along-coast variation in the rate of mean tide level rise may reflect time of deglaciation, neotectonics, or differential isostatic adjustments. Between 8 and 5 ka (cal), only south-central Maine (Phippsburg) has a good record of relative sea-level change. At this locality, the rate of mean tide level rise was 5.0-8.8 mm/yr for the period 7.8-5.3 ka (cal), which may have resulted from collapse of a glacial forebulge. A slight acceleration of mean tide level rise has occurred during the past millennium in Gouldsboro and Machiasport. If 12 m downwarping in easternmost Maine occurred, as suggested in other publications, it must have happened prior to 5.7 ka (cal).