Page:Africa by Élisée Reclus, Volume 3.djvu/21

 THE AZOEIAN WATERS. 5 drawn between the continental masses owing to llie gradual transition of relief, geology, and climate, still less can any well-defined lines of demarcation be traced across the oceanic basin. Following the motion of the sun along the ecliptic, the system of aerial and marine currents is subject to incessant modifications. The seasons are alternately displaced from north to s6uth, and from south to north, while the ever-restless floods, setting now in one, now in another direction, inter- mingle the climatic zones in their ceaseless changes. It must suffice to indicate in a general way as the dividing zone the relatively narrow section of the Atlantic comprised between the submarine plateau of Western Europe and the Bank of Newfoundland. This is the " telegraphic " bed, the first part of the ocean that has been sj^stematically explored for the purpose of laying the cables between Europe and America. The bed of this region, which has a mean depth of over 2,000 fathoms, presents the greatest regularity, the most gradual slopes and uniform depths for vast tracts in the whole marine basin. The section which stretches south of the telegraphic plateau, and which may be called the Azorian Atlantic, from the archipelago lying nearest to the centre, is sufficiently, well defined southwards by the narrower zone comprised between Africa and South America. A line drawn from the Bissagos Archipelago through the islets of Saint Paul and Fernam de Xoronha to Cape Saint Roque, has a length of less than 1,750 miles, and in this dividing zone the waters are somewhat shallower than in the sections lying to the north and south. In its general outline the Azorian Atlantic forms a flattened crescent with its convex side facing westwards, and bounded by the United States, the Antilles, the Guiana and Brazilian coasts. The inner concave line is indicated by the African seaboard from the Strait of Gibraltar to Cape Palm as. The bed of this section of the Atlantic is much more irregular than that of the North Atlantic to the south of Greenland and Iceland. The oceanic depths between Africa and the Antilles are not only interrupted by several insular groups and the plateaux supporting them, as well as by the surrounding banks, but in this region there also occur numerous submarine mountains, which, like the upheaved archipelagoes themselves, are very probably due to lava formations. These submarine masses are met chiefly to the west of the Azores, where a sudden subsidence of 3,000 feet would reveal the presence of numerous islands, some disposed in scattered groups between the telegraphic bed and the Azores, others stretching from this archipelago for some fifteen degrees of longitude westwards in the direction of the Bank of Newfoundland. Most of the banks, however, indicated on the old charts — those, for instance, of Sainte-Marie and Kutusov — south of the Azores, have not been met during more recent soundings. Earthquakes, waterspouts, schools of cetaceans, or floating masses of pumice, have often led to the assumed existence of reefs in places where the plummet descends to depths of 14,000 to 16,000 feet without touching the bottom. Sadden changes of colour, from green to blue, from blue to black, usually correspond in the Azorian Atlantic to varying depths. Such at least is the inference drawn from these coincidences by the members of the Tallyman