260 



DEPARTMEls'T OF THE ^'AYAL SERVICE 



layer, but the warmest water in the tank was that which flowed from the warm centre 

 to the cold. 



XI 





iif 



f.^".-. 





,l::.(^n^^7^:^j^::^^y,;~^:^-:,rg,: 



I 



<'!■■'« 



I 





x: 



w 



_±L 



xJ 





Fig. 44.— Ex|)eiiment illustrating circulation of the Gulf Stream. 



A further experiment was thereafter made with both warm and cold centres placed 

 at the surface of the water. When stationary conditions had been reached, it was here 

 found that a very slight degree of circulation was taking place between the warm and 

 the cold centre, the entire remaining mass of water in the tank being motionless, as 

 shown in fig, 44 e. Finally, the cold centre was placed at a lower level than the warm. 

 In this case, as with the foregoing experiments, a highly intensive circulation at first 

 arose throughout the whole of the tank, but as soon as this had given place to a 

 stationary situation, it was seen that the entire mass of water in the tank remained 

 altogether motionless. The water at a higher level than the lower side of the warm 

 centre had then the same temperature as the warm centre itself, and that below the 

 level of the top of the cold centre the same temperature as this latter, vide fig, 44 d. 



