NATURE OF THE STREAM 137 



but there are practical problems to be overcome in using it. 



One cannot sight across an ocean in order to measure the 

 slope of the water, but an ingenious Dane, N. E. N0rlund, 

 solved this problem recently by laying pipes across the Great 

 Belt between the Danish islands of Zealand and Fyn, a dis- 

 tance of about twelve miles, and across the Sound between 

 Denmark and Sweden, a distance of three miles. Water in the 

 pipe comes to the same level at each end, except for tempera- 

 ture and other effects which may be compensated for, and 

 thus a basis of comparison is provided for sea level measure- 

 ments. 



The Danish pipe arrangement, while perfectly feasible 

 over a short distance in fairly shallow water, is beset with 

 formidable difficulties in longer distances and deeper waters. 

 For this reason the direct measurement of sea level has not 

 hitherto been widely applied in the calculation of ocean cur- 

 rents, and the Atlantic oceanographers have used a less direct 

 but more practical application. The displacement of lighter 

 and heavier water layers across the width of the ocean cur- 

 rents, related to the water slope, can be found by lowering 

 thermometers and steel sample bottles to different depths 

 at a series of stations across the current. The density or heavi- 

 ness of water depends both on its temperature and its salinity, 

 both of which are measured by the deep-sea instruments, 

 and therefore the modern oceanographer uses them indirectly 

 to measure current. 



The accuracy of this method was dramatically shown by 

 Wuest in 1924. From temperature and salinity measurements 

 between Miami and the Bahamas made on the Blake in 

 1878 and the Bache in 1914, he calculated that a current of 

 over three knots flows at the center of the Stream and that 

 at some places a current extends almost to the bottom. What 

 was of greater interest was that the calculated currents were 



