The Northeast Atlantic passive margins formed during the early Cretaceous phase of rifting. From Goban Spur to Galicia Bank, these margins were characterized by a tilted fault block geometry, well defined by numerous single-channel and multichannel seismic data. Stress directions cannot be deduced from these structural data, but recently Withjack and Jamison (1986) have described the strain states and fault patterns produced by oblique rifting using experimental and analytical techniques. We have applied these techniques to the Northeast Atlantic continental margins. All types of extensional margins are represented there with pure extensional rifting (Goban Spur and West Galicia margins), oblique rifting (North Spanish and Celtic margins), and close to pure wrenching (Armorican margin). The rifting directions are parallel to the adjacent oceanic fracture zones or flow lines defined in the Northeast Atlantic and in the Bay of Biscay, except for the West Galacia margin where a 20¿ difference exists. This difference could be within the error range of the method or could be due to a slight relative motion between Galicia Bank and Iberia during rifting. Globally, the coherence between rifting and spreading directions is an argument in favor of the validity of the method. On the basis of kinematic arguments, the motion of Iberia with respect to Eurasia has an eastward component related to the west to east motion of Africa with respect to Eurasia, while Iberia was a small plate whose motion was roughly coupled to Africa. This is confirmed by our analysis. We conclude that during both the rifting and spreading episodes in the Bay of Biscay, the motion of Iberia with respect to Eurasia was in the southeastward direction. Recently, Boillot (1984) suggested that Iberia moved about 150 km southwestward with respect to Eurasia during the rifting phase and then moved in the southeastward direction during the spreading phase. We suggest that the Boillot hypothesis could be wrong. ¿ American Geophysical Union 1989