214 DISCOVERY REPORTS 



(see Table VII, p. 170). Subsidence between the dates June 25 and July 8 and August 

 7-20 remains a probability but cannot be regarded proven. 



SINKING OF NEWLY MIXED WATER 



Indications of this phenomenon off Caldera (p. 140) leads one to suppose that the 

 mixing of upwelled with oceanic water must lead frequently to the formation of heavier 

 water which thereupon sinks on the outer edge of the zone of uprising. Such sinking of 

 the product of recent mixing is, of course, to be distinguished from subsidence of 

 recently upwelled water. Upwelling and subsidence of the lower layers may be de- 

 scribed as a see-saw motion dependent upon variations of wind or of other forces re- 

 sponsible for the divergence of surface water from the coast: subsidence restores 

 equilibrium by the passive method of letting the water layers revert towards a condition 

 of horizontal stratification. Sinking of heavier water, on the other hand, restores equi- 

 librium by disrupting thermal stratification; it is irreversible in its action. Although 

 newly mixed water which is sinking is independent of any external force other than 

 gravity, it may be to a large extent dependent on wind action for its inception. Thus 

 at Caldera its appearance coincided with the convergence of warm, offshore water with 

 the cool inshore water, and with local northerly wind. Whether the cool water inshore 

 was welling up concurrently with this convergence of the warm water, as a result of 

 the southerly winds recorded offshore or of winds even farther from the coast, or 

 whether the cool inshore water was subsiding, is not known. 



Other instances of sinking of newly mixed water are probably to be found among 

 our sections; though most to be expected in the neighbourhood of the subtropical 

 convergence, it is probably a phenomenon of importance in the mixing process of the 

 coastal current waters. At Cape Carranza, an appearance of sinking water at St. WS 

 597 is also correlated with northerly wind, but cannot illustrate this phenomenon since, 

 south of the subtropical convergence, the water is less saline offshore than inshore. 



CENTRES OF UPWELLING AND. OTHER IRREGULARITIES 



OF THE CURRENT 



CENTRES OF UPWELLING 



Schott has suggested that the length of coast between Coquimbo and Punta Parina 

 may be divided into four regions,^ each distinguished by strong upwelling which sets in 

 with a sudden lowering of temperature and then dies away northwards so that finally 

 almost normal temperatures are found over short distances. In Schott's graph com- 

 paring inshore temperatures with those at 80-100 miles offshore, the points of con- 

 spicuous upwelling are Antofagasta, San Juan and Punta Aguja. The points of least 

 upwelling are north of Coquimbo, south of Antofagasta, Arica and Puerto Chicama. 



^ I have used the word " region " instead of Schott's word " zone " because I have already used the word 

 zone in a different sense elsewhere. 



