1870.] 



Deep-sea Researches. 



211 



II. That so long as the difference of level is maintained, so long will 

 this flow continue ; and thus any agency which permanently keeps the level 

 of one body of water below that of the other (unless it directly antagonize 

 the downward pressure of the higher water*), will maintain a permanent 

 surface-flow from the higher towards the lower. This constant tendency 

 to equalization will keep the actual difference of level within very narrow 

 limits. 



III. That wherever there is a want of equilibrium arising from difference 

 of density between two columns of water in communication with each other, 

 there will be a tendency towards the restoration of equilibrium by a flow 

 from the lowest stratum of the denser column towards that of the lighter, 

 in virtue of the excess of pressure to which the former is subjected. 



IV. That so long as the like difference of density is maintained, so 

 long will this flow continue ; and thus any agency which permanently dis- 

 turbs the equilibrium in the same sense, either by increasing the density 

 of one column, or by diminishing that of the other, will keep up a perma- 

 nent flow from the lower stratum of the denser towards that of the less 

 dense. — This constant tendency to restoration of equilibrium will keep the 

 actual difference of density within definite limits. 



V. That if there be at the same time a difference of level and an 

 excess of density on the side of the shorter column, there will be a ten- 

 dency to the restoration of the level by a surface-ftow from the higher to 

 the lower, and a tendency to the restoration of the equilibrium by an under- 

 flow in the opposite direction from the heavier to the lighter column. 



VI. That so long as the difference of level and the difference of 

 density are maintained, in the same sense, so long will each flow continue ; 

 and thus a vertical circulation will be kept up by any continuous agency 

 which alters at the same time both the level and the density of the two 

 bodies of water, — provided that the excess of density is on the side of the 

 lower column. 



VII. That the rate of each flow, where it is not confined within defi- 

 nite limits, will depend simply upon the amount of disturbance, in the one 

 case of level, and in the other of density ; and when this disturbance is 

 small, it may be so slow as to be almost imperceptible, though not less 

 real and effective. But if the communication between the two bodies of 

 water take place through a long narrow channel, the rate of movement will 

 increase so as to produce a decided current in each direction j since the 



* Thus it has been shown by Archdeacon Pratt, that in consequence of the local 

 attraction produced by the high land of Asia, with nothing but Ocean to the southward, 

 the sea-level at the mouth of the Indus is no less than 515 feet above that at Cape 

 Comorin (Philosophical Transactions, 1859, p. 795). So, again, if Barometric pres- 

 sure be lower over any Oceanic area than on other parts of the surface, there will be an 

 elevation of the water-level in that area, equilibrium being reached when the excess of 

 Water-pressure becomes equal to the deficiency of Air-pressure. (See Mr. T. Gc. Brent 

 in Philos. Transact. 1867, p. 5.) 



S2 



