226 DISCOVERY REPORTS 



SOUTHERN BOUNDARY 



Authors of text-books usually look to the West Wind Drift for the origin of the Peru 

 Current (Hoffmann, 1884; Somerville, 1923). The West Wind Drift is described as 

 dividing into two branches at its point of impingement (usually given as 40° S) against 

 the South American continent : the Cape Horn Current flowing to the south, the Peru 

 or Humboldt Current to the north. 



It is probably true that the greater part of the sub-Antarctic water which turns north- 

 ward is derived from the West Wind Drift. Such water may be considered as belonging 

 to the Peru Current in the wider sense, and particularly to the Peru Oceanic Current. 

 It remains oceanic until it has mixed with upwelled water. For the southern boundary 

 of the Peru Coastal Current therefore, we should look for the most southern upwelling 

 centre. The southernmost limit has been variously set by different writers at Cobija 

 (Vallaux, 1930), Copiapo (Deutsche Seewarte), the region between Antofagasta and 

 Coquimbo (Schott) and Valparaiso (Hoffmann, 1884). Schott records that upwelling 

 began somewhat north of Coquimbo in about 29° S in September 1927. He points out 

 that this satisfies the theory of a correspondence between depression of the temperature 

 of the coastal waters and a low rainfall on the adjacent land. Incidentally it is close to 

 where the subtropical convergence approaches the coast: i.e. where upwelled water 

 becomes less saline than the open ocean. Coquimbo has a rainfall of about 100 mm. 

 whereas Taltal in lat. 25-5° has about 15 mm. and comes definitely within the upwelling 

 region. While Schott makes this point in his account of the Peru Current, it is to 

 be noted that according to the chart published by Schott and Schu (1910) surface iso- 

 therms of the Pacific curve northwards near the coast in lat. 40° S. Since any northerly 

 drift of the surface water is likely to entail some upwelling the temperature at the 

 commencement of the Peru Coastal Current is liable to be lower than at the commence- 

 ment of the Cape Horn Current, as was observed by Perez-Rosales (1857).^ 



Our own observations show that upwelling may occur at least as far south as Cape 

 Carranza and with strength off Copiapo. Upwelling is no doubt liable to occur at any 

 point off the coast where northerly drift happens. Mossman (1909) has shown with 

 some precision that the dividing line between the meteorological cyclonic and anti- 

 cyclonic circulations normally lies at about 41° S. South of this the prevailing wind is 

 west-north-west in all seasons, whereas northwards the prevailing wind alternates with 

 season. From October to March it is southerly ; from April to September northerly ; it 

 brings the monsoon to the Chilean littoral, and may be felt as far north as Caldera. 

 Hoffmann (1884) states that between Valparaiso and the 40th parallel, a northerly gale 

 in winter is always followed by a strong southward current. Gunckel (1928) states that 



^ According to Murphy (1923) "the current laves the western coast of South America from a point some- 

 where south of 40° S". But he also says: "It is in accord with the abyssal topography that the current as 

 characterized by upwelling waters, should begin near Valparaiso, for just south of that port the great bank on 

 which Juan Fernandez lies extends to the westward more than twenty degrees of longitude." We have shown 

 on p. 205 that the Juan Fernandez Rise has no appreciable connection with the surface waters. 



