126 



KEPORT — 1870. 



There were very sensible values of A^, B^ of series L for all years, and also 

 of A^, Bj of series ; biit the resulting amplitudes and corrected epochs did 

 not agree, showing (as theory gives) no tides of these periods. The cause of 

 the apparent irregularities of series L is probably satisfactorily explained 

 hereafter. 



56. Shortly before the Meeting of the British Association at Exeter in 

 1869, application -was made to Graham H. Hills, Esq., for a Liverpool Tide 

 Diagram-sheet containing curves influenced by a minimum amount of ir- 

 regularity due to wind, atmospheric pressure, temperature,, tfec, and which 

 might be taken as practically representing the astronomical tide-curve. 

 Through his kindness a sheet containing a period of thirteen days (1869, 

 April 24th to May 6th) was received, and has proved of great value as a 

 practical test of results and (§§ 61-63) demonstration of new tide-compo- 



