WAVE OF HIGH WATER. 307 



run in the manner of ordinary waves in canals, he says 

 that its great and important defect is, that the water is not 

 distributed over the surface of the globe in canals of uniform 

 breadth, or in any form very nearly resembling them. In 

 this regard its fundamental suppositions are probably as 

 much, or nearly as much, in error as Laplace's theory ; 

 but it masters the peculiarities of river tides, which no other 

 theory has touched upon. 



This ample confession of the imperfections of existing 

 causal theories must plead our excuse for offering a few 

 hints towards a new theory of the tides. These hints are, 

 however, merely incidental to the subject of this paper, 

 which is practically limited to the consideration of the 

 law of direction of the progressive motion of the wave of 

 high water. 



Our present knowledge of the progress of tidal waves is 

 derived from observations of the time of high water on 

 ocean coasts — the direction in which the hours increase 

 along a given coast being held to indicate the direction of 

 the progressive motion of the wave. The theories of Newton 

 and Laplace would lead to the conclusion that tidal waves 

 should tend to follow the direction of the moon's motion 

 from east to west ; but observation has shown that the real 

 progression is nearly everywhere at right angles to that 

 direction, the hours increasing from north to south, or 

 south to north, indifferently, on the shores of the principal 

 land-areas of the globe. This anomalous progression has, 

 in fact, rendered those theories of little or no value when 

 applied to the consideration of the probable direction of 

 the progressive motion of tidal waves. 



In 1833 Dr. Whewell published his first map of co-tidal 

 lines. His method of grouping the facts of tidal hours is 

 based upon the supposition that the observed succession of 

 those hours on ocean coasts was due to the progression of 

 free waves of translation, whose origin was to be traced to 



x2 



