290 REPORT— 1876. 



The tides at Port Leopold and at Beecliey Island, as seen from the above 

 analyzed constituents (due allowance being given to the different values for 

 I), are almost identical in character, the agreements of the components "when 

 compared being remarkably close. 



The diurnal components are large, and in the j'ears of maximum declination 

 of moon nearly sufficient at certain parts of the lunation to reduce the period 

 of the tide to that of one tide only in the twenty-four hours. 



The narrrowness of Behring's Straits precludes the supposition that the 

 tides in the Arctic Ocean and among the channels of the Arctic Archipelago 

 are sensibly influenced by communication through it with the Pacific. 



The largeness of the values for the retardation of spring-tides renders it 

 probable that for these places the tidal influence is derived chiefly from the 

 Atlantic Ocean, and not by direct action of Sun and Moon on the waters of 

 the Arctic Ocean. 



The shortness of the series of these observations, especially those for 

 Beechey Island, has rendered some departure from the usual mode of reduction 

 desirable. The following explanation is therefore necessary :— The observa- 

 tions were first grouped according to mean solar hours, and the summations 

 and means obtained and the means analyzed. The values of the mean solar 

 amplitudes thus obtained were then used for the calculation of the height of 

 the tide at each integral mean solar hour due to these constituents. The 

 heights were then copied for the previous values grouped according to mean 

 solar hours, subtracting from each quantity the effect of the height due to 

 the mean sun, computed as above explained. These numbers were grouped 

 according to mean lunar hours, and the series of mean values analyzed as 

 before. The value of the tide at each mean lunar hour was then computed 

 and subtracted from the previously copied series, and grouped according to 

 sidereal hours and the reduction continued as before ; and so on throughout 

 the seven evaluated series. The values obtained for the smaller components 

 found by these means are no doubt trustworthy. 



Notes on the Reductions of the Tidal Observations of Brest and Toulon. 



By Mr. Roberts. 



"With the kind assistance of M. Janssen the tracings from the original 

 " registered tidal observations at Toulon for the whole of the year 1853 were 

 " obtained from the Department of the Marine at Paris, and also a copy of 

 " the heights at Brest for the year 1875. These two years' observations have 

 *' been treated and reduced in a similar manner to the many years' tidal obser- 

 " vations fully described in the Eeports of the Tidal Committee of the British 

 " Association for the years 1868 to 1872. It wiU therefore be only necessary 

 " in this place to give a general summary of the results obtained. The ob- 

 " servations being given in metres, the results have been similarly expressed 

 " in decimetres and centimetres, and not in English feet. 



" Brest. 



" The proportion between the solar and lunar semidiurnal tides at Brest is 

 " (allowing for the value of the Moon's declination) about as 1 to 3, and 



