1915- No- 2. SPITSBERGEN WATERS. 



57 



equal to that ot the ice. If we assume that the icefloe, to which the ship 

 was moored (see Fig. 35), had an average thickness of about 4 or 5 metres, 

 it seems probable that its movement should represent the avera<ye move- 

 ment of the upper layer — about 5 metres thick — of the sea; and there 

 was no wind during the period of observation, which could influ.nce the 

 drift of the ice appreciabh-. 



As some parts of the ice may descend deeper, it mav also be in- 

 tluenced by the movement of the water ac this greater depth; but this 

 must be of much less importance than the effect of the current of the top 

 layer of the water upon the main body of the ice. On the other hand it 

 must, however, be remembered that a heavy ice-floe, such as that to which 

 the ship was moored, represents a great mass which has a considerable 

 vis ine/iiœ, and cannot therefore be expectd to follow at once all chan^^es 

 in the movements of the water of the top layer of the sea. It seems there- 

 fore possible that the effect of the tidal wave may often turn the direction 

 of the water somewhat more rapidly than that of the ice. Thus the direc- 

 tion of the movement of the ice and the top layer of the sea mav not 

 coincide at all moments, especially when there are rapid changes in the 

 direction. 



We may nevertheless expect, however, that the movement of the ice- 

 tloe represents on the whole the average movement of the top laver of 

 the sea. although irregularities may, now and then, anse, especiallv, per- 

 haps, from the pushing of the ice-floes against each other during the drift, 

 as was mentioned before. 



It would have been of much importance to the studv of the above 

 question, if our observations at 10 metres could have given fairly trust- 

 worthy information about the movement of the water at that depth; but 

 unfornateh" the observations indicate that the movements often change so 

 suddenly and so irregularly that we should want a much larger number 

 of observations, taken at short intervals, in order to be able to trace the 

 movements at this depth in their details. 



Our curves of velocity and direction. Figs. 36 Å: 37. and our central 

 vector diagram. Fig. 38. representing the movement at 10 metres, give an 

 idea of the great irregularities. But even these curves had to be construc- 

 ted hypothetically for long intervals; and if there had been more obser- 

 vations the curves would certainly have been much more irregular. An ex- 

 ample of the irregularity of the movement at this depth is given, for in- 

 stance, by the observations .between 7.21 and 7.31 a.m. on August i8h, 

 when, during the ten minutes the observation lasted, the direction of the 

 movement at 10 metres changed 160 '\ from X4o'^\V to S 60'' E, and this 



