The opening of the Central Atlantic Ocean basin that separated North America from northwest Africa is well documented and assumed to have started during the Late Jurassic. However, the early evolution and the initial breakup history of Pangaea are still debated: most of the existing models are based on one or multiple ridge jumps at the Middle Jurassic leaving the oldest crust on the American side, between the East Coast Magnetic Anomaly (ECMA) and the Blake Spur Magnetic Anomaly (BSMA). According to these hypotheses, the BSMA represents the limit of the initial basin and the footprint subsequent to the ridge jump. Consequently, the evolution of the northwest African margin is widely different from the northeast American margin. However, this setting is in contradiction with the existing observations. In this paper, we propose an alternative scenario for the continental breakup and the Mesozoic spreading history of the Central Atlantic Ocean. The new model is based on an analysis of geophysical data (including new seismic lines, an interpretation of the newly compiled magnetic data, and satellite derived gravimetry) and recently published results which demonstrate that the opening of the Central Atlantic Ocean started already during the Late Sinemurian (190 Ma), based on a new identification of the African conjugate to the ECMA and on the extent of salt provinces off Morocco and Nova Scotia. The identification of an African conjugate magnetic anomaly to BSMA, the African Blake Spur Magnetic Anomaly (ABSMA), together with the significant change in basement topography, are in good agreement with that initial reconstruction. The early opening history for the Central Atlantic Ocean is described in four distinct phases. During the first 20 Myr after the initial breakup (190-170 Ma, from Late Sinemurian to early Bajocian), oceanic accretion was extremely slow (similar to 0.8 cm/y). At the time of Blake Spur (170 Ma, early Bajocian), a drastic change occurred both in the relative plate motion direction (from NNW-SSE to NW-SE) and in the spreading rate (an increase to similar to 1.7 cm/y). After a small increase between Chron M25 (similar to 154 Ma, Kimmeridgian) and Chron M22 (similar to 150 Ma, Tithonian), the spreading rate slowed down to about 1.3 cm/y and remained fairly constant until Chron MO (125 Ma, Barremian-Aptian boundary). In addition, kinematic reconstructions illustrate a significant spreading asymmetry during the early history of the Central Atlantic Ocean; the accretion rates were higher on the American side and led to the formation of more oceanic crust on this plate. We infer that this asymmetry could be related to the fact that the thermal anomaly responsible for the significant magmatism of the Central Atlantic Magmatic Province (CAMP) was preferentially located below the African plate.