t 



Ijf* 



TIDES. 



[Sect. IV, 



lighest for the high water, and note the height and the 



fj TV> ^ 



U.I 



; and in like manner for low water. 

 6. But the following is a better mode of dealing with 

 observations thus made every five or ten mmutes. Let a 

 number of parallel lines (ordinates) be drawn at intenals 

 corresponding to the intervals of observations, and bomido.i 

 by a line perpendiculai- to them on one side (the abscissa), 

 and on these lines (the ordinates) let the observed height^ 

 of the surface be set off (from the abscissa) and let a line 

 be drawn through the extremities (of the ordinates). This 

 line, if it be tolerably regular, will give the time of hidi 



watei 



i» * 



and if it be somewhat irregular, it 



can be 

 curve, and then the time and hei'dit of 



smoothed into 



bdgh water read off. And in like manner for low vrater. 

 Suppose, for example, that we have the following ob- 



servaiions of the 

 minutes for an hour : 



height of the water made 



every At 



Times of Ob- 

 servation . 



SO ! 35 i 40 : 45 ; 50 55 



m. 

 60 



Heights ob- 

 served 



11 11 



6 

 11 



6 

 9 



6 

 5 



6 



! 



5 

 10 



The selection of the greatest height (as in Art. 5) 



would give high water at Oh. 30ra. : but the general run 



of the height (Art. 6) would give the high water two or 



three minutes later, as appears by drawing the dotted 

 curve in fig. 1. 



7. It IS easy to draw such curves, if we have, ready 

 prepared, jmjjer rvlod into small squares, the divisions in 



the horizontal line reprcbenting hours and minutes, and 



\ 



I 



I 



( 



r 



[ 



1 



^ 



Sect 



IV 



!^ -' 



r 



ijji 



the d 



£ 



8. 



any I 

 the w 

 quart 

 irreg] 

 from 

 that i 



t 



mijst: 



izi, 



case 







every 



s 



case 

 perio 



\ 



