440 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1956 



a satisfactory system. A remarkable feature of this method is that 

 under some conditions the measurement of water currents may be 

 carried out without even leaving the shore. Since the electrical field 

 caused by currents extends beyond the edge of the water the electrodes 

 may be used on land where they measure the speed of water from the 

 terrestrial part of the electrical field. 



Water in the open sea is not exactly the same from top to bottom. 

 As mentioned previously, sea water is lighter when it is warmer and 

 fresher, and heavier when it becomes cooler and saltier. As might be 

 expected, when undisturbed by currents or mixing processes, the sur- 

 face layers of the sea will be lighter and successively deeper layers 

 will be increasingly heavy. The exact density can be calculated from 

 the temperature and salinity. When a current flows, however, there 

 is a readjustment of the distribution of density in the water to com- 

 pensate for the earth's rotation, which exercises an effect on moving 

 bodies known as the Coriolis force. The effect of this is to shift the 

 heavier water toward the left of the current when looking down- 

 stream in the Northern Hemisphere. The degree of this shift is pro- 

 portional to the current. The oceanographer can therefore calculate 

 the flow of a water current at sea provided he knows the way in which 

 the water density is distributed. 



For practical purposes, the research vessel steams at right angles 

 to the current, stopping at intervals to make the necessary measure- 

 ments. At each station a cable is sent down with a number of water- 

 sampling bottles attached to it at appropriate intervals. The Nansen 

 bottles, as they are called, are made of steel tube, and are sent down 

 with both ends open so that the water runs through them. When each 

 arrives at its proper depth, messenger weights are sent down the 

 cable so as to trip a trigger mechanism that turns the bottle upside 

 down on its hinged attachment to the cable, while at the same time 

 closing it. The temperature is meanwhile measured by means of a 

 sensitive thermometer, attached to the bottle, which automatically 

 records the temperature registered at the time it is upended. The 

 saltness of the water is measured by chemical analysis of the sample 

 brought back in the bottle. 



The thermometer used is guarded against the pressure of water by 

 being enclosed in a protecting tube. A second thermometer is car- 

 ried on the Nansen bottle for the surprising purpose of measuring 

 the depth at which the sample is taken. In order to do this, this ther- 

 mometer is not provided with a protecting tube. As a result, the water 

 pressure causes the bulb to be slightly compressed and so the thermom- 

 eter registers higher than it should. The amount of pressure and 

 therefore the depth of water can be calculated from the difference in 

 reading of the two thermometers. 



