174 DISCOVERY REPORTS 



highly saline water at the surface, bounded both inshore and offshore by water of lower salinity, shows 

 that the stations have sectioned the northern edge of oceanic surface-water which was involved in the 

 large eddy lying south of this line. The isotherms and isohalines both slope upwards towards the 

 coast at the inshore stations, but this does not necessarily mean that active upwelling was taking place. 

 The same effect could be created by the presence of the oceanic water so close to the coast, compressing 

 the coastal type of water into a narrow strip adjacent to the land. The salinity and temperature 

 distribution in such a strip of water would then necessarily, on account of the laws governing the 

 density distribution, resemble the pattern created by upwelling. 



The isolation of this inshore water from the deeper water, apparent in the salinity section (Fig. 30), 

 suggests that active upwelling was not taking place. 



To the north of this section another centre of upwelling may have been present at Cape Frio (see 

 Fig. 7 b) in 18 30' S., but there are insufficient data to decide this question. 



To sum up, it is evident that the same essential features were present on both of these surveys. 

 On the second survey we can distinguish the same two types of surface water. 



(a) The coastal or upwelled water, which on survey II is characterized by temperatures below 

 14-0° C. and salinities of less than 35-00%,,. The convection layer in this water is better developed 

 than on survey I, but the underlying discontinuities are not so pronounced. These differences must 

 be caused by the stronger winds and upwelling on survey II. 



(b) The offshore water, characterized by temperature over 14-0° C. and salinities of over 35-10 % . 

 This is in general cooler, but very similar in other respects to the same type of water on survey I. 



In survey II, active upwelling was found on the two southern sections, and possibly also on the 

 Walvis Bay line, while only on the northernmost section (the Mowe Point line) were quiescent condi- 

 tions encountered. 



Table 6. Stability of the 0-50 m. and 0-100 m. layers 



Survey I Survey II 



