64 FRIDTJOF NANSEN. M.-N. Kl. 



movement would have a tendency to be deflected towards the right both 

 upwards and downwards from that layer, if no other forces have any 

 influence upon the direction. 



According to the measurements between August i8th 3 p. m. and Au- 

 gust 19th 2 p. m. (see p. 59, and Fig, 45), the direction of the current at 

 100 metres was towards N67*'E, or 10^ to the right of the current at 

 50 metres, and the velocity at 100 -metres (2.6 cm. per second) was lower 

 than that at 50 metres (4.8 cm. per second). The measurements between 

 2 a.m. and 2 p.m., August iSth, give the following values of the currents 

 according to the progressive vector diagram Fig. 44: 



At the Surface 12.6 cm. /sec. towards S 74OW 

 » 20 metres 10.5 » » N25°W 



» 50 » 1.3 » » N68^E 



» 100 » 5.4 » » N 20° E 



Hence we find that in those periods when the current at 100 metres 

 had a higher velocity than that at 50 metres, its direction deviated towards 

 the left of the latter, while it deviated towards the right when the velocity 

 was lower. This would be in agreement with our theory. If, however, we 

 take the mean values given by the measurements between August i8th 

 2 a. m. and August 19th 2 a. m. we get somewhat different results (see 

 Fig. 44). 



The observations that, at 10 metres the water probably formed a 

 reaction-current, moving in a direction about opposite to that of the sur- 

 face drift, makes, the whole current-system still more complicated in 

 our case; and it is not probable that the direction of the current at 20 

 metres is directly influenced by the direction of the surface-drift. It seems 

 more probable that the direct effects of the surface-drift was limited to the 

 nearly homogeneous top layer with low salinity and density, a little more 

 than 19 metres thick. This layer should thus have a current-system of its 

 own, with an average surface-drift towards about Sôô^W, in its upper 

 part, and a current of compensation in an opposite direction in its lower 

 part, about 10 metres; and there may even be many complications as to 

 direction and velocity inside the small current-system of this top layer. 



Below the light surface-layer there was warmer and more saUne water 

 which must evidently have come from the south. This is also borne out 

 by our current-measurements, according to which the water at 20 metres 

 was on the average flowing very nearly towards true north, with a mean 

 velocity of about 8 cm. per second. The velocity of this current probably 

 decreased with increasing depth, while its direction changed more and more 

 towards the right down to some depth below 50 metres. Still deeper, the 



