Along the Atlantic coast of Morocco a series of embayments or basins including the Essaouira-Jebel Amsittène and Tarfaya-Laayoune basins, contain marine sediments which may attain considerable thickness and whose Early Jurassic (Liassic) age has been well documented. This paper describes Liassic deposits from six exposures or drill holes between the Saharan coast of Tarfaya and the Essaouira-Jebel Amsittène region. These Atlantic basins do not connect with the marine troughs of the High and Middle Atlas, which follow a Mediterranean Tethys trend, but are separated from them by a SSW-NNE trending threshold made up of the axial portion of the central Hercynian Massif (Moroccan Meseta). Their opening to the ancestral Mediterranean must have been farther north, probably by means of interconnecting, WSW-ENE trending rift-related graben trenches, possibly starting with the Rharb (Kenitra) basin. More likely, however, openingsexisted beneath the present nappes of the Rif and along the undoubtedly wide fracture zone represented by the Azores-Gibraltar line separating the Rif terrain from that of the Betic Cordilleras (both of which contain well dated marine Hettangian). Farther SW, probably hidden by the sedimentary prism of the African continental shelf, connections may exist with the Liassic occurrence of Central America, as already suggested by Avias (1953. Sci. Terre 1 (1), 1–276; 1956. 20° Congr. géol. Intern Mexico, Secc. II, 1–5), shown in a figure by Erben (1956b. Neues Jb. geol. paläontol. Abh., Stuttgart 103, 28–79), and subsequently upheld by Hallam (1971a. J. Geol. Chicago 79 (2), 129–157; 1983. Palaeogeogr. P. clim. P. ecol. 43, 181–193) and Thierry (1982. Bull. Soc. géol. Fr., Sér. 7, 24, 1053–1067). These lines of communication would have centered on a “Panamanian Strait” (“Panama Strasse”) and would have permitted faunal migrations, in particular towards South America. Migrations of certain European and Mesogean faunal species have recently been proposed by Schmidt-Effing (1976a. Münster Forsch. Géol. Paläont. 38–39, 201–217; 1976b. Publ. geol. ICAITI, Guatemala, 5, 22–23; 1980. In: The Origin of the Gulf of Mexico and the Early Opening of the Central North Atlantic Ocean (Edited by Pilger, R. H., Jr), pp. 79–86, Von Hillebrandt (1981b. Geol. Rdsch. 70 (2), 570–582; 1984. Int. Symp. Jurassic Stratigraphy 3, 716–729) and Riccardi (1983. In: The Phanerozoic Geology of the World (Edited by Moullade, M. and Nairn, A. E. M.), II. B, 201–264). Unless one believes that in all localities cosmopolitan faunas arrived from a universal ocean, these relationships lead to the model of an “Atlantic corridor”, starting in the Middle, and perhaps even the Early Liassic (the “Paleotethys” of Bernoulli and Lemoine, 1980. Mém. Bur. Rech. géol. min. 115, 168–179; or the “proto-Atlantic” or “Atlantic Tethys” of Lancelot, 1980. Mém. Bur. Rech. géol. min. 115, 215–223; and Lancelot and Winterer, 1980 Initial Report of the Deep Sea Drilling Project 50, 801–821; although the latter authors postulate such a connection only from Late Liassic time on). This model raises the specific question whether or not Early Jurassic deposits are present in the Senegal basin (Guieu, 1976. Rapp. Dépt. Géol. Fac. Sci. Univ. Dakar 32, 1–87).