700 ■ REPORT— 1894. 



devised by Mr. Clayden, in which water disposed over an area shaped like the 

 Atlantic, and sprinkled over with lycopodium dust to make movement apparent, 

 was subjected to air impelled from various nozzles, representing the mean direc- 

 tions of the permanent winds. It dispelled the last doubt I held on the subject, as 

 not only were the main currents reproduced, but the smaller effects and pecu- 

 liarities of the Atlantic drifts were produced with surprising accuracy. 



There is a small current, long shown on our charts, but which I had always 

 regarded with suspicion. I refer to the stream which, after travelling from the 

 Arctic Ocean southward along the east coast of Greenland, turns sharply round 

 ■Cape Farewell to the northward into Davis Straits, where it again doubles sharply 

 on itself to the southward. This is exhibited, in the model, in all its details, and 

 is evidently caused by the pressure of the water forced by the mimic Gulf Stream 

 into the Arctic region, where it has no escape except by this route, and is pressed 

 against the land, round which it turns as soon as it can. This is, no doubt, the 

 «xplanation of the real current. 



The very remarkable winter equatorial current, which runs in a narrow belt 

 eastwards, just north of the main stream travelling west, was also reproduced with 

 extraordinary fidelity. 



The winds, however, that are ordinarily considered permanent vary greatly, 

 ■while in the monsoon areas the reversal of the currents caused by the opposite winds 

 exercise a great influence on the movements of the water far beyond their own limits, 

 and anything like a prediction of the precise direction and rate of an oceanic stream 

 can never be expected. 



The main facts, however, of the great currents can be most certainly and simply 

 explained in this manner. 



The trade winds are the prime motors. They cause a surface drift of no great 

 velocity over large areas iu the same general direction as that in which they blow. 

 These drifts after meeting and combining their forces eventually impinge on the 

 land. 



They are diverted and concentrated and increase in speed. They either pour 

 through passages between islands, iis into the Caribbean Sea, are pressed up by 

 the land, and escape by the only outlets possible — as, for example, the Strait of 

 Florida, and form a great ocean current like the Gulf Stream — or, as in the case of 

 the Agulhas current and the powerful stream which runs north along the Zanzibar 

 coast, they are simply pressed up against and diverted by the land, and run along it 

 with increased rapidity. 



These rapid currents are eventually apparently lost in the oceans, but they in 

 their turn originate movements of a slower character, which on again passing over 

 Bhallow water or on meeting land develop once more into well-defined currents. 



We find an analogous state of things on the western side of the Pacitic, where 

 the Japan current is produced in a similar manner. 



The fact that on all western shores of the great oceans towards which the trade 

 winds blow we find the strongest currents running along the coast, is almost enough 

 of itself to prove the connection between them. 



The westerly winds that prevail in higher northern and southern latitudes are 

 next in order in producing great currents. From the shape of the land they in 

 Gome ca«es take up and continue the circulation commenced by the trade winds ; in 

 others they themselves originate great movements of the water. 



Compared to the great circulation from this source the effect of differences of 

 temperature or of specific gravity is insignificant, though no doubt they play their 

 part, especially in causing slow under-circulation, and in a greater degree the 

 vertical mixing of the lower waters. 



No drop of the ocean, even at its greatest depth, is ever for one moment at rest. 



Dealing with minor points, the American ofiicers of the Coast and Geodetic 

 Survey have found after long and patient iuvestigation that the velocity of the Gulf 

 Stream in its initial and most marked part, the Strait of Florida, is greatly affected 

 by the tide, varying as much as one-half its maximum rate during the twenty-four 

 hours. 



These American investigations are of greatest interest. They have extended over 



