CANADIAN FISHERIES EXPEDITION, 1914-15 259 



The effect of the currents originating in physical change in the water is then 

 simply that of transporting water of a certain character from a region where such 

 water abounds to regions where it is rare. And the quantity of water carried by such 

 a current will depend entirely upon the production and consumption of the kind of 

 water in question. The Gulf Stream carries about 25,000,000 tons of water per second. 

 Hence, it follows that a corresponding amount of such water is produced in the tropics 

 and consumed in the Arctic regions at the same rate. 



And from this we may also conclude, that the complementary physical processes 

 whereby water is added to or drawn from a certain layer adapt themselves so as to 

 balance one another. What takes place is evidently this : a decrease of the quantity 

 of water in a layer favours the development of forces tending to introduce water from 

 without, and vice versa. And the present state of the water throughout the sea is the 

 result of thousands of years of such adaptation. 



The Archimedean forces which drive a current of this nature adapt themselves to 

 the quantity of water to be carried forward and the degree of resistance to be encoun- 

 tered. In order to understand this, we may once more take the Gulf Stream as an 

 example. If the production of Gulf Stream water in the tropics, and corresponding 

 consumption of the same in the Arctic regions were to cease, then the lower boundary 

 surface of the Gulf Stream would soon become horizontal, and the Archimedean forces 

 which drive the current forward would disappear, with the result that the forward 

 movement of the water itself would cease. If, on the other hand, the production and 

 consumption of Gulf Stream water were to become greater than is at present the case, 

 then the lower boundary surface of the Gulf Stream would assume a still steeper slope, 

 and the Archimedean forces attain a higher degree of intensity than at present. This 

 increase of force would arise in order to meet the necessity for transport of a greater 

 volume of water than the Gulf Stream is now required to carry. 



In order to demonstrate the importance of the Sargasso vortex to the Gulf Stream, 

 the following experiment was made. A rectangular tank, 50 cm. long, 50 cm. deep, and 

 5 cm. broad, and having glass walls, was filled with water, and heating and cooling 

 apparatus iiitroduced, the latter constructed on the same principles as applied to the 

 heating of rooms, etc., i.e., consisting of metal receptacles through which a current of 

 warm or cold water could be transmitted. The pipes through which the heating or 

 cooling water was carried to the receptacles were carefully isolated. In accordance 

 with the conditions actually met with in the ocean, the cold centre was placed at the 

 surface of the water at one end of the tank, and the warm one a little below the surface 

 at the other end. The cold centre thus represents the cooling of the water in the Arctic 

 regions, the heating apparatus answering to the corresponding influence at the bottom 

 of the Sargasso vortex. 



After the experiment had been in progress for some time, and stationary condi- 

 tions arrived at, some potassium permanganate solution was introduced into the water 

 near the middle of the tank, and immediately beneath the surface. The solution, 

 easily visible on account of its colour, moved over to the cold centre, and then sank to 

 the lower level of the warm centre, moved across to this, and then rose thence to the 

 surface of the water, recommencing here its movement towards the cold centre again, 

 as shown in fig. 44 a. Owing to the force of the circulation, the circulating layer of 

 water soon became coloured right through, until only a narrow strip running obliquely 

 down remained untinged, vide fig. 44 b. At last this, too, disajipeared, and the entire 

 layer was coloured. The water layer beneath the warm centre remained entirely 

 uncoloured, thus indicating that no interchange takes place between the circulating 

 layer and that subjacent. On introducing a similar coloured solution into this lower 

 layer, it appeared that the movement of the water here was altogether insignificant. 

 The temperature of this layer was the same as that of the cold centre itself. The cur- 

 rent leading from the cold to the warm centre was slightly warmer than the bottom 



105573 



