210 DISCOVERY REPORTS 



temperature changes off Palominos Island may be similarly related to the position of 

 convergence of the wedge. 



Swirls of the type described must continually vary in size and location in accordance 

 with distant and local forces : and this should be borne in mind when weighing evidence 

 of correlation between a wind which was light and the upwelling off Palominos Island, 

 or again between a wind which is very local such as the virazon and the convergence of 

 warm water with the Pisco-Callao-Guanape Islands stretch of coast. 



The possible working here of a seiche is discussed on p. 212. 



Other localities 



Other instances of an apparently direct relation between surface temperature and 

 wind have already been noted; of cool inshore water and south-easterly wind at 

 Pichidanque Bay (p. 140) and San Juan (pp. 148-51), and warm inshore water with 

 north-westerly wind off Cape Carranza (p. 135), and in the Caldera neighbourhood 

 (p. 140). Compare also the seasonal changes noted on pp. 226-7. 



At many localities, however, there was an appearance of upwelling in comparatively 

 calm weather. The evidence at Pichidanque Bay, for example, goes to show that con- 

 ditions had been calm for many weeks before our observations were made; both 

 nutrient salts and plankton were depleted and a thermocline was becoming established 

 at 40-50 m.: yet traces of definite upwelling were present (Fig. 21). Again upwelling 

 off Callao and perhaps off the Guanape Islands seems heavy for the strength of local 

 winds. At Arica there was no sign of an earlier meteorological disturbance : calm weather 

 on this part of the coast is traditional, and out to sea a thermocline at 30-40 m. was 

 clearly established (Fig. 31). Yet upwelling from a depth of 50 m. was conspicuous. 



Evidence of a foregoing period of calm, at Arica and Pichidanque Bay at any rate, 



precludes the possibility of interpreting the upward curve of isotherms and isohalines 



as subsidence on an extensive scale. These must be examples of upwelling caused by 



forces at some distance from the regions under consideration. Such was indeed inferred 



by Dinklage in 1874 (Schott, 1891, p. 215) and by Buchan (1895). The latter writes: 



It is probable that the great volume of, and distance travelled by, these currents in the broad Pacific 

 as compared with other oceans directly results in a stronger and more widespread upwelling, accom- 

 panied with a correspondingly extensive diminution of temperature. 



The formation of currents by aspiration is well known and is well illustrated off 

 Northern Peru where the coastal water off Punta Aguja and Punta Parina is drawn 

 west-north-west in the wake of the South Equatorial Current (Ferrel, i860, p. 55). 

 The strong inshore current of 48 miles a day off Arica, and, in part, the inshore current 

 off the Guafiape Islands, may be supposed to be due to the same cause, for each of 

 these localities lay to the southward of regions of strong surface drift. 



It becomes a question whether a coastal current caused by aspiration in this manner 

 may not diverge from the coast as a result of the earth's rotation and so induce up- 

 welling. In respect of wind-induced current, Ekman concludes: 



The most striking result of the coast's influence is that a wind is able indirectly to produce a current 

 more or less in its own direction from the surface down to the bottom, while in the absence of coasts the 

 wind's effect would be limited to a comparatively thin surface layer. 



