Segmentation is a characteristic feature of seafloor spreading along the global mid-ocean ridge system. While segmentation of active spreading centers has been the focus of much recent research, the process by which a rifted continental margin develops into a segmented mid-ocean ridge is still poorly understood. In this study we investigate the segmentation character of the U.S. East Coast margin through a modeling study of the margin basement structure, magnetics, and gravity anomalies. The East Coast margin is of particular interest because it is one of several rifted continental margins that display thick sequences of high seismic velocity igneous crust, presumably formed during rifting. The East Coast Magnetic Anomaly (ECMA), a distinct total field magnetic high running offshore along the margin, is commonly located seaward of the thickest sections of the high-velocity crust and displays segmentation on length scales (100--120 km) similar to the segmentation observed at the Mid-Atlantic Ridge (MAR). Isostatic gravity anomalies were calculated by removing from free-air gravity the predicted effects of seafloor, sediments, and a crust-mantle model assuming local isostatic compensation. The resultant residuals show a corridor of high anomaly running along the margin, situated close to the maximum thickness of the high seismic velocity crust as determined from the two available seismic refraction lines. Reduction to the pole (R-T-P) of the total field magnetic anomaly shows that after the removal of skewness from the ECMA, the location of the isostatic gravity high is closely correlated to the ECMA. The isostatic gravity high is also segmented but in two distinct wave bands: 100--150 km and 300--500 km. The short-wavelength (100--150 km) segmentation in the R-T-P magnetic and isostatic gravity anomalies is similar in wavelength to segmentation in magnetization and mantle Bouguer anomaly observed along the present-day MAR. The 300--500 km segmentation in the along-margin isostatic gravity anomaly is similar in length scale to both intermediate-wavelength tectonic segmentation observed in the South Atlantic and variations in lithospheric strength observed along the African margin. Furthermore, two of the intermediate-wavelength (300--500 km) isostatic gravity lows correspond to the early traces of the Kane and Atlantic fracture zones, suggesting that these two fracture zones may define boundaries of a single tectonic corridor in the North Atlantic. We hypothesize that the direct cause of the intermediate-wavelength segmentation may be along-margin variations in both the amount of underplated igneous crust and the strength of the lithosphere, although the relative importance of these two effects remains unresolved. Our results imply that segmentation is an important feature of margin development and that segmentation at mature oceanic spreading centers may be directly linked to segmentation during continental rifting. ¿ 2001 American Geophysical Union