20 The Geostrophic Relationship 



very small differences of very large numbers. It is difficult to place much 

 confidence in them. 



Thus the choice of the reference level for geostrophic calculations be- 

 comes mostly a matter of taste, and we should admit that that is ultimately 

 intolerable. The determination of the level of no motion is not a matter 

 for debate, but for direct measurement — a subject to which I shaU return 

 in a polemical section at the end of this book. 



Fig. 1. Schematic cross section of the Gulf Stream. The arrows indicate 

 direction of horizontal pressure gradient. The little circles indicate vanishing 

 horizontal pressure gradient. The current is flowing perpendicular to the plane 

 of the page. The dashed lines are contours of equal water density. The line I 

 is a level surface; the line s is the actual sea surface. 



SCHEMATIC DENSITY AND PRESSURE FIELDS 

 ACROSS THE GULF STREAM 



Fig. 1 is a schematic diagram showing a hypothetical cross section of the 

 Gulf Stream. A level surface (such as the ocean at rest would assume) is 

 shown by a heavy solid line. The hypothetical sea surface across the Gulf 

 Stream is showTi by the thin soHd fine. The difference in elevation between 

 one side of the Stream and the other is supposed to be about 1 m. The 

 vertical scale of the diagram is exaggerated very greatly to show the shape 

 of the free surface. The broken fines beneath the surface represent contours 



