For a quarter of century, it as been recognized that the opening of the Atlantic ocean, the opening of the Bay oi Biscay and the evolution of the Pyrenees are part of the same question. Understanding the three series of events thus requires a global approach. Yet from a mass of ever better constrained date and opinions have arisen more misunderstanding and more contradictions than has coherent synthesis. The reason for this lies in an excess of fragmentary or one-sided conclusions. From an analysis of geological and geophysical data and a review of published kinematical reconstructions, this study aims at having an in-depth look at presently conceivable solutions. First are considered the concepts oi plate, blocks, intra-plate deformation, first and second order plate boundaries, kinematical models and kinematical phases; and then how to estimate the precision of the reconstructions is discussed. The evolution oi the Iberian plate is in the larger context of the main plates that surround it - Eurasia, North America and Africa - as well as of the smaller ones which play a role in the opening of the North Atlantic and in the evolution oi the Arctic : Greenland, Rockall, Svalbard. This evolution includes that of the present ocean floor checked by analysis of the magnetic anormal record and also an estimation of the intra-continental deformation, linked either to formation intra-continental bassins and continental margins, or to peri-Atlantic mountain building, Furthermore, the reconstructions concern the evolution of the whole area since as early as th Late Hercynian age. The conclusions are presented with respect to three separate questions : 1. the origin of the Pyrenean area. In terms si two widely accepted hypothesis -(a) the attenuation of the continental crust beneath the bassins and margins would results exclusively from a stretching process and (b) the deep peri-Iberian bassins (the Tagus abyssal plain, the Armorican bassin, etc.) would have opened in Jurassic and Cretaceous time - we propose a detailed solution (reconciling the movements oi a detain of plates or blocks) which is coherent with these from a kinematical point oi view. But we lay stress on the numerous difficulties or impossibilities for such a solution from a geological point of view, which render questionable the initial hypotheses, 2. the evolution of the Pyrenean area during the opening of the Bay of Biscay. We show that it is possible to model the evolution of the Pyrenean area between chrons MO and 34 in two phases : opening between 120 and 107 Ma and strike-slip between 107 and 90 Ma. This model is compatible with both the Atlantic kinematical data and the pyrenean geological data, 3. the Pyrenean evolution during the compressional phases. We show that the total compressional movement between the Iberian meseta and Europa cannot be smaller than 150 km.