202 DISCOVERY REPORTS 



The breadth of the zone of actual uprising is extremely narrow compared with the 

 breadth of the zone influenced by upwelled water, but upwelling does not always seem 

 to be in immediate contact with the coast. The breadth of the region of upwelling may 

 be placed variously from 5 miles at Punta Aguja to may be as much as 30 miles off'shore 

 near the Guanape Islands (Figs. 35 and 37). An example of upwelling at some distance 

 from shore was found at Antofagasta (Figs. 26 and 28) and possibly also at Pichidanque 

 Bay and at Capo Blanco (Figs 21 and 41). At Capo Blanco a small patch of water of 

 less than 17° C. was crossed at a distance of 31-35 miles from shore; it had a breadth 

 of 4 miles and lay between the tongue of the Equatorial Counter-current inshore 

 and water of the Peru Current south and west. In this area of mixing and of local 

 eddies it would be interesting to know whether this cool water had come to the surface 

 as the result of an eddy in the open sea or had its origin under the coast farther to 

 the southward (see also reference to Schott's divergence line, p. 228). 



Indirect upwelling by a process of vertical mixing is to be expected not only in the 

 zone of actual uprising but for many miles westward wherever wind is heavy and the 

 thermocline not too pronounced (Atkins, 1924), and this adds to the problem of deter- 

 mining the limit of the zone of actual uprising. 



EFFECT OF DIRECTION OF COAST-LINE 



Hydrologists have demonstrated by theory (Ekman, 1905) and experiment (Sand- 

 strom, 1 91 9) the principles that underlie upwelling phenomena, but as Ekman has him- 

 self stated, conditions in the sea are so complicated that it is impossible to calculate 

 exactly the motions of the ocean. He has therefore taken a number of type problems in 

 which some factors as they occur in nature — such as the shape of the ocean basin, the 

 winds and the distribution of temperature and salinity — have been replaced by simplified 

 imaginary ones. The principles Ekman has demonstrated may be regarded as tendencies 

 which sea water in movement will show under various conditions : but conditions in the 

 sea diff'er so materially from those postulated in the type problems that any comparison 

 between Ekman's theories and our findings should be drawn with caution. Sandstrom 

 has emphasized the dynamic importance of isosteric surfaces across which movement 

 of the water is checked but along which it is facilitated. This report does not seek to 

 detail water movements beneath the surface with exactitude, and the isosteric surfaces 

 are not determined. The west coast of South America presents a number of problems, 

 and various suggestions have been put forward to explain them. It is not inappropriate 

 to discuss some of these in the light of the results obtained. 



After remarking that in 1927 and 1929 the depression of temperature proper to the 

 Peru Current is greater off Peru than off Chile, Schott states that the boundary between 

 the two regions lies near Arica, that is where the coast bends suddenly. He states 

 further that during the 'Emden's' cruise this boundary was particularly well marked; 

 with high (normal) temperature and high salinity upwelling seems to have been nearly 

 or quite absent. His explanation runs as follows : " Es muss hier infolge der veranderten 

 Kiistenrichtung fiir eine grossere oder kleinere Strecke die Stromrichtung zum Land 



