i8 95 THE DISTRIBUTION OF FOOD-FISHES. 35 



the addition of large quantities of fresh water from rivers, is never- 

 theless on account of its low temperature — due to land influences — 

 markedly denser than the oceanic waters outside. As spring pro- 

 gresses and the atmosphere becomes warmer, the surface layers of the 

 North Sea gradually rise to a temperature higher than that found in 

 the ocean ; and as they thus become specifically lighter, they tend, 

 as we have already seen, to flow outwards. The warm surface layers 

 at the same time protect the lower strata from being warmed by the 

 atmosphere, and the latter retain their winter cold, except in so far as 

 they are warmed by conduction. In early summer we have, there- 

 fore, at the meeting place to the north of the Shetlands, the 

 following conditions : a cold under-layer of North Sea water, 

 somewhat fresher than the oceanic water, but denser than it 

 because of its lower temperature ; a warm layer of North Sea water, 

 fresher and lighter than the oceanic water, but tending to flow out 

 over it ; and lastly the oceanic water, Salter than either of the others, 

 but intermediate in density because of its temperature. Under these 

 circumstances a gradual process of mixture is set up, with the result 

 that between the oceanic and bottom North Sea water a ridge of 

 maximum density or heaviest water is formed, due to the saltness of 

 the one and the cold of the other. In the earlier stages the position 

 of the ridge is probably nearly north and south from the extreme 

 northern point of the Shetlands ; but as the season advances and 

 mixture progresses, the axis gradually retires eastwards. Its position 

 in August, 1893, is shown by the dotted lines in the chart (Fig. 1), 

 the salinity being as represented by the solid lines. 



As soon as the axis of maximum density has moved eastwards 

 clear of the Shetland Islands, it is evident that a path is open for 

 oceanic water to move southwards between it and the land, as a 

 bottom current. Guided by the ridge, the water will penetrate gradu- 

 ally southwards along the east coast of Scotland ; and as the process 

 of mixture goes on, the ridge itself will slowly retire, and oceanic 

 water spread more and more to the east into the North Sea. As 

 winter approaches, the falling temperature and increasing wind will 

 gradually destroy this distribution, and cut off further influx from 

 the ocean until the following season. 



Obviously the conditions required to bring about the circulation 

 just described are exceedingly complex. The amount of oceanic 

 water admitted to the North Sea in this way, and the date of its 

 admission, must depend on the meteorological conditions prevailing 

 in the north-east of the Atlantic, in the southern part of the Norwegian 

 Sea, and in the North Sea itself, during at least six months pre- 

 viously ; and it may be that in cold and stormy seasons some entirely 

 different circulation takes place. The observations of the " Jackal," 

 however, leave almost no doubt that the movements of water de- 

 scribed actually took place in the summer of 1893. Probably the 

 unusually fine season, and the great cold of the preceding winter, gave 



D2 



f s 



