1915- No. 2. SPITSBERGEN WATERS. 43 



drifting, and the direction of the drift was indicated by the direction of 

 the vertical plane of the sloping sounding-line. Theoretically this method 

 should give fairly accurate results., but practically this is hardly the case. 

 Owing to its weight, the sounding-line will form a curve, and the greater 

 its angle with the vertical, the greater does the curve become. It will 

 consequently be necessary to pay out more line than would have been 

 the case i( the line could have been kept straight; and it is hardly- pos- 

 sible to calculate how much. Owing to this circumstance the method will 

 therefore give too great values for the drift of the ship. On the other 

 hand there is naturall}- a risk that the lead may be dragged along the 

 bottom, and if so this will naturally tend to make the mearurements of the 

 drift too small. Another drawback with this method is the difficulty of 

 measuring the direction of drift with sufficient accuracy, and it is especially 

 difficult to measure the changes in the direction of the drift during the 

 time of the observations, because the direction (/. e. the azimuth) indicated 

 by the Une will change much more slowely than the direction of the drift, 

 especially if the line forms a great angle with the vertical, and there is a 

 great distance between the ship and the lead on the bottom. K the line 

 is kept tight it will point the whole time towards the lead, and even though 

 the direction of the drift may change a great deal, it can cause only a 

 verv slow alteration in the direction of the line. The currents at inter- 

 mediate depths may also have appreciable effect upon the direction of the 

 sloping Une. This kind of obser\ation is not therefore of much value for 

 computing the actual current at higher levels from obserxations made simul- 

 tan eoush* with a current-meter from the drifting ship. 



The method might, however, have been usefiil for controlUng the 

 obser\ations of the drift of the ship made by a current-meter lowered to 

 the water-strata near the bottom. But unfortimately this was not done 

 regularly, as we had too few winches with sounding-line, and too few men 

 for the observations. After having made several attempts with the above 

 method at Stations 19 and 20, on August 4th and 5th, and after having 

 foimd by computation that it did not give satisfactory' results, I gave up this 

 method, and decided instead to work with two current-meters, one lowered 

 to a water-stratum near the bottom, and the other used simultaneously at 

 higher levels. Some few observations were, however, made simultaneously 

 with the two methods. 



On .A.ugu3t 5th, from 4. 11 to 4.21 p. m. an obser\ation vas made with 

 a current-meter at 540 metres. The depth of the sea was then 575 metres. 

 The obser\-ation gave no movement, and at that time and during the pre- 

 vious hour the obser\-ations also gave no movements at hisrher levels. It 



