THE WATER MASSES 183 



The eddy which lies to the north of the Orange river line is probably a fair representation of the 

 westerly flow of coastal water adduced from the temperature and salinity observations. 



On survey II the movements are considerably more intense. Undoubtedly the sparsity of data 

 shows a simpler picture than is shown in the surface-temperature chart (Fig. jb) which is constructed 

 from a much greater number of observations, but within the three great eddies shown, the cooler 

 coastal waters show a pronounced offshore, or longshore movement to the north and west. In 

 conjunction with this the oceanic waters move in an easterly direction towards the coast. 



There were very few occasions on which the set of the ship could confidently be ascribed to water- 

 movement as distinct from wind leeway. At the station on the outer end of the Walvis Bay line on 

 survey I a northerly set of about 1 knot was encountered with a light wind. This agrees with the 

 computed currents (Fig. 34). The only other occasion upon which we can be reasonably certain the 

 effect was solely due to currents was between stations WS 1057 and 1061 on survey II. When the ship 

 was steaming north from the Orange river line stations to the inshore end of the 25 ° S. line, she was 

 set 065 ° at h knot with no wind. Once more this is in agreement with the computed currents. 



Topography of the 200 db. surface 

 The lesser number of observations at this depth give the appearance of a very simplified pattern. 

 On survey I there is clearly a southerly movement along the coast from the north, penetrating south- 

 ward and tapering in towards the coast, and finally disappearing in about 28 S. Seawards of this the 

 water appears to move to the N.N.W. 



On survey II offshore the principal movement is once more to the north, but there is some indica- 

 tion of a south-easterly flow on the landward side of a trough line running north-west through 

 WS 1070. This tendency to a southerly or onshore movement at this depth in the water adjacent to the 

 continental shelf will later (p 1 90.) be seen to be connected with the process of upwelling. It is postulated 

 that this is a compensatory movement replacing the water which has been drawn up to the surface 

 inshore and we have, therefore, called it the compensation current. 1 This southerly current is warmer 

 and more saline than the waters seaward of it in the trough which separates it from the northerly 

 current farther offshore. 



The water masses at 200-600 m. 

 Generally speaking, on both surveys, the waters in this layer correspond very closely with the mean 

 temperature-salinity relationship already outlined for the South Atlantic central water. Table 8 gives 

 the salinity values corresponding to the given temperatures at each station. On the first survey the 

 mean range of salinity at the given temperature is only o-n % , and on the second survey o-io% . 



Table 8. Temperature-salinity relationships of the water masses between 200 and 600 m. 



Survey I Survey II 



The salinity values, corresponding to the given temperatures at each station are shown in parts per thousand. Mean: Survey I = o-ii%„; 

 II = 0-10%,. 



1 Yoshida and Mao (1957), and later workers have shown that divergence and consequent upwelling at the surface is 

 likely to be accompanied by a poleward movement in the subsurface layer. 



