338 DR ALEXANDER BUCHAN ON 



movements. It may be remarked that, while the general depth is measured by two or 

 three miles, the extent is measured by thousands of miles. 



Attention may be drawn to a singular circumstance. While in the South Pacific, to 

 the east of long. 140° W., none of the observed salinities are above the average to the 

 south of lat. 33° S., in the South Atlantic the separating line is about lat. 40° S., and 

 in the Indian Ocean, lat. 43° S. The differences are probably due to the regions in the 

 different oceans of the great ice-meltings that take place. 



Leaving out of view the movements of the waters of the ocean caused by winds, all 

 other movements, whether vertical or horizontal, must be referred to differences of specific 

 gravity. But while this is so, it is seen that -for movements of great bodies of waters 

 which have actually been effected, the maps of salinity furnish equal if not more autho- 

 ritative information as to how the transference has been effected from one region to 

 another, or from one depth to another. This arises from the extensive serious blanks 

 which occur in all parts of the ocean, rendering it impossible in many cases to trace these 

 movements in their progress from one region to another. 



In the South Atlantic it is seen that at all depths the temperature and salinity are 

 very greatly lower off the coast of Africa, whence the prevailing winds blow, than off the 

 coast of South America, towards which they blow. Further, that the layers of warm water 

 at or near the surface, which are everywhere little more than a mere film as compared with 

 the mass of cold water beneath, are very greatly thinner on coasts where the wind blows 

 seaward from the land, than off the coasts on which the prevailing wind blows home. 

 This peculiarity holds good in all parts of the globe. In the one case cold water upwells 

 to supply the drain on the surface waters by the winds driving them from the coasts, 

 and in the other the prevailing winds drive the warm surface waters before them, and 

 thus cause an accumulation of warm water of very considerable thickness along the 

 coasts they strike. In this connection Dr Murray has pointed out that there are 

 no coral reefs off' the western shores of Africa or of America, because they have 

 not there the high temperature they require. They are found, however, off the 

 lee shores of tropical and sub-tropical regions, whither prevailing winds transport the 

 waters of high temperature on which the existence and well-being of the corals 

 depend. 



There is a different type of upwelling in the Pacific, Atlantic, and Indian Oceans, 

 which is best examined through the temperatures.* The most important of these is in 

 the Pacific, and may be briefly described. Within the area of this remarkable upwelling 

 the " Challenger " took serial temperatures in lat. 7° 35' N. and long. 149° 49' W. In con- 

 trast with these, other serial temperatures were observed across the equator in lat. 11° 20 

 S. and long. 150° 30' W., this point being on the eastern side of the high temperature and 

 high salinity area to the east of Australia. The following extracts from the "Chal- 

 lenger" soundings, Nos. 410 and 417, taken respectively on August 10th and September 

 14th, 1875, will show the extraordinary differences in the rates of change of temperature 



* For a more detailed account of these see the Challenger Report, p. 19. Consult also Maps 3 and 4 of that Report. 



