74 BULLETIN OF THE NATURAL HISTORY SOCIETY. 



(3) Time of Low Water at Spriuyhill. From Table I it is 

 seen that low water at Springhill occurs on an average eleven 

 hours and nine minutes later than low water at St. John. Hence 

 it takes low water one hour and of rty-nine minutes longer to 

 travel from St. John to Springhill than it does high water. This 

 is shown in another way by the shape of the Springhill curves. 

 It will be noticed that in all cases the curves are steeper on one 

 side of high water than on the other, the tide rises faster than 

 it falls, so that a low water alwaj's comes closer to the succeeding 

 high water than to the preceding. In fact, the average time 

 from low water to high water is only five hours and seventeen 

 minutes, while that from high water to the next low water is 

 seven hours and seven minutes. This relative delay of low water 

 is due to one of the diflerences between wave motion in a shallow 

 river and wa\e motion on the ocean. In the former the more 

 elevated parts of a wave always travel faster than the less eleva- 

 ted or the depiessed parts. In fact, if v lie the velocity of any 

 part of a wave whose elevation above the mean level is h, and if 

 // be the depth of the river 



In this, r is of course the value of r, for parts of the wave for 

 which h is zero ; that is, for parts of the wave midway between 

 crest and trough. In the parts of the wave below mean water 

 level h is negative. Hence v is greater for the crest than for the 

 trough ; that is, gi^eater for high water than for low water. Thus 

 low water keeps lagging farther and farther behind the high 

 watei- ahead, and a])i)roaching the high watei' behind. This 

 process may go so far that the front of the tide wave becomes 

 nearly vertical and then we have a bore as in the Seine, Petit- 

 codiac and many other livers. 



With this greater steepness of the front of the tide wave 

 another peculiarity is often developed. The rear slope of the 

 wave may first becoiiic sttaiglit and tlu-n actually recurved. 

 This is hardly shown in any marked degree on the St. John 

 River, although the rear slope sometimes approximates to a 



