202 



DISCOVERY REPORTS 



but its origin is uncertain. Part of it upwells far south and mixes with the fresh water 

 from melting ice and snow to form the Antarctic surface water; part of it becomes 

 cooled but only slightly diluted and gives rise to the heavy Antarctic bottom water. 

 Both waters have a northward movement. There is then in the Antarctic Zone a move- 

 ment of cold water away from the pole at the surface and bottom and a movement of 

 warmer water towards the pole at an intermediate depth. 



Subtropical surface water is water which has spent some time circulating in sub- 

 tropical or tropical regions. Near the subtropical convergence it appears to have either 

 a component of movement towards the south or a smaller northward movement than 

 the sub-Antarctic surface water. 



There are generally sudden increases of temperature, and sometimes of salinity, as 

 one crosses the convergences between the zones from south to north. The increase in 

 temperature across the Antarctic convergence is 2 or 3° C: where the convergence is 

 far north (50° S) it is from about i to 3-5° C. in winter and from about 3-5 to 6° C. in 

 summer ; where the convergence is far south (60° S) the figures are o to 2-5° C. in winter, 

 2-5 to 5° C. in summer. In the Pacific Ocean and in the western part of the Atlantic Ocean 

 there is a small increase of salinity across the convergence from south to north : from 

 33-8 or 33-9 %o to 34-1 or 34-2 °l^^; the few observations in the eastern part of the 

 Atlantic Ocean and the Indian Ocean make it appear that there is no increase there. The 

 position of the convergence appears to be fixed by the deep-water movements, and so 

 far as is known it varies only within very small limits. 



The subtropical convergence is well defined in the Atlantic Ocean and the western 

 part of the Indian Ocean. It is usually marked by a sudden increase of temperature of 

 about 4° C, and of salinity of about 0-5 %o. It is not so well defined to the south of 

 Australia and in the Tasman Sea. There have been very few observations in the Pacific 

 Ocean, but the convergence appears to be well marked there except in the east. There 

 have not been enough observations to give a full account of the temperature and salinity 

 changes across the subtropical convergence, but it can be said that its position is 

 approximately that of the 11-5° C. isotherm in winter, of the 14-5° C. isotherm in 

 summer, and of the 34-9 °j^^ isohaline. It does not appear to be stationary everywhere 

 like that of the Antarctic convergence. It seems to vary most in the Atlantic Ocean and, 

 so far, a definite seasonal variation has been found only to the south of the Brazil 

 current: there its average position is 3-5° farther south in summer than in winter. 

 It has other variations which are not seasonal changes. 



Briefly the surface temperature limits of the Antarctic, sub-Antarctic and sub- 

 tropical Zones are as follows: 



