TIDAL PHENOMENA OF THE ST. JOHN RIVER. 71 



high water at 8 a. m.; the range from high water to low water 

 being four and a half inches. The instrument was then removed 

 to be slightly altered, and at S p. m. of July 28th it was replaced 

 at the same point, and thenceforward a continuous record was 

 olitained until August Gth. This would have given eighteen high 

 waters and seventeen low watei's, were it not for the fact that 

 the weather, which, until the .31st, was calm, suddenly became 

 stormy, with winds of as much as twenty miles an hour from the 

 north-west, that is, from nearly exactly up-stream. These were 

 sutiicient to totally obliterate the ordinary tidal rise and fall and 

 give curves whose ragged irregularities represented faithfully 

 every variation in speed and direction of wind. In the complete 

 record which accompanies this I may point especially to 7 p. m. 

 of the -Ith, 8 a. m. of the .")th, 1 p. m. of the 30th, and 1 a. m. of 

 the 2nd. The corresponding wind velocities, kindly supplied by 

 Dr. Harrison, have for comparison been placed at the top of the 

 record. A curious hump in the curve at 10 p. m. of the 5th was- 

 explained by a sudden gust of wind which, Dr. Harrison informed 

 me, was indicated by the recording wind gauge just at that time. 

 These facts are interesting as indicating the very great effect 

 which wind has on water in a somewhat confined basin. The 

 effect would, of course, be still more marked in the case of lakes. 

 This is of interest in connection with tlie other paper (on second- 

 ary undulations) presented to the Society. 



The smoothness of the curves in calm weather is of import- 

 ance as indicating that we have at Hpringhill no true forced 

 vil)ration of the water, produced directly by lunar influence, but 

 only a free derived waved started by the rise and fall at St. John. 

 A mixture of l)oth would give irregularities in the cuive. 



VII. Time and Amount of Hioii Water at Springhill. 



In the curves obtained at Springhill, there are in all twelve 

 fairly well marked high waters, the others being unreliable on 

 account of wind disturbances. Of these, six were obtained during 

 \'ery calm weather and six others during windy weather. From 

 the former, Table I has been calculated. Column 2 gives the 



