No. 2. SPITSBERGEN WATERS. 69 



As was pointed out above, the currents at all depths vary so much 

 from hour to hour as to velocity and direction, that it is doubtful whether 

 the inclinations of the strata may correspond chiefly to the average velo- 

 city and direction of the current, or whether these inclinations may change 

 so rapidly that they are more in accordance with the actual current at 

 every moment. The latter possibility seems, however, less probable, especi- 

 ally if the current is broad, because it would necessitate so considerable 

 a transport of water that there would hardly be time for it. 



By taking only the current-measurements made between August i8th 

 3 p. m. and August i9th 2 p. m. we found above (p. 59) the following 

 average values for the currents : 



At the Surface 12.6 cm. /sec. towards S 62 W 



10 metres 1.9 N 44 E 



20 5.6 N 8 E 



. 50 4.8 N 57 OE 



roo 2.6 N67E 



If these angles are introduced, our computations will giwe the fol- 

 lowing values: 



At the Surface 13.1 cm. /sec. towards S 62 W 



10 metres 0.9 N 44 E 



20 [51.9] N 8 E 



50 4-3 N 57 OE 



100 2.5 N 67 E 



This agreement is also remarkably good, except of course at 20 metres. 



Current Measurements on August 4th and $th. 



As was mentioned above, the current measurements made at Stations 

 19 and 20, on August 4th and 5th, were very incomplete, chiefly on ac- 

 count of the very small number of observations made in the deep layers 

 near the bottom, by which the drift of the ship, moored to the ice, could 

 be determined. During this time there were also many difficulties with the 

 ice, which made it necessary to look after the ship; and the number of 

 observations were thus reduced. As was also mentioned above, several 

 attempts were made to determine the drift of the ship by means of the 

 sounding-line and the lead lying on the bottom; but this method proved 

 unsatisfactory. 



Figs. 56 to 59 represent the curves of velocity and direction of the 

 relative movements at the different depths, as far as it was possible to 

 construct them with the incomplete observation material. Some of them 

 are rather hypothetical, as there were too few observations. This is especi- 



