TIDAL PHENOMENA OF THE ST, JOHN RIVER. 81 



begun to fall, it is still running up river. At low water the 

 converse happens, that is, the level at Indiantown keeps falling 

 nearly until the part below the rapids has been filled up to the 

 level above the rapids. It seems, however, not a little remark- 

 able that the delay is so exactly the same for high water and 

 low water. This account must be admitted to be very imperfect, 

 as I had very little time for exact observations, except as 

 regards the time of high water and low water at Indiantown. 



XIII. — Tidal Currents. 



In the preceding I have not paid any attention to the currents 

 which form so important a part of tidal phenomena. The sub- 

 ject is one of great complexity in such a river as the St. John. 

 Two remarks may however be made. 



First it is rather a common mistake to suppose that there 

 must exist a flow of saltish water as far up the river as tides can 

 be detected. Two grounds are sometimes advanced for this view. 

 The first is that a tide means a flow of water and there must be 

 a flow of water as far up as there is a tide. This statement is 

 true, but the deduction is unsound ; for a tide, whether in a river 

 or on the ocean, is a wave, and a wave may pass on for thousands 

 of miles while the water at any place only makes short excursions, 

 going forwards as the crest of the wave passes by, and backward 

 as the trough passes. No one would claim that the water at the 

 mouth of the Bay of Fundy travels the whole way to the head 

 with the tide, for if so a vessel could float that distance in one 

 tide. The second ground sometimes advanced is that there must) 

 by the principles of hydrostatics, be salt water as far up steam 

 as the point at which the bed of the river is on a level with the 

 mean level of salt water at the mouth of the river ; and that hence 

 up to the head of the tides there will be an undercurrent of salt 

 water up and an overcurrent of fresh water down. But it 

 is impossible that two such layers should co-exist for a hundred 

 miles without mixing. Again in many rivers such as the Ama- 

 zon, La Plata, and Forth it is known that the tides extend a 

 long distance further up than the point at which the level of the 



