ON THE COASTS OF IRELAND. 9 



less certainty than could be desired; at the same time I have no hesitation in ex- 

 pressing my belief, that the credit of the results hereafter to be given is not sensibly 

 injured by this circumstance. At Glenarm, from a similar cause, it became neces- 

 sary to refer to the post-office clock ; but the observations do not appear to have 

 suffered materially. 



The observations began about June 22, and were discontinued about August 22. 



Section II. — Methods of extracting from the observations the times of high and low 

 water ; of supplying deficient times and heights ; and of correcting the times first 

 determined. 



The determination of the height of the water, at high or low water, from the 

 observations, was a matter of no difficulty. In two or three instances of low water, 

 when the water had dropped below the zero of the tide-gauge, the observations were 

 incomplete till it again rose to the zero ; in these, the observations were supplied by 

 comparison with other low waters which had been completely observed. 



The determination of the time was far more difficult. The examination of these 

 observations has made me very distrustful of the results which have been deduced 

 from observations of time only. The difficulty of fixing on the precise time of high 

 or low water will appear from this statement, that sometimes twenty or twenty-four 

 successive observations (occupying 1^ 40°^, or 2^') are registered with the same decimal 

 of a foot for the height. The most perplexing case is that where the change of height, 

 in respect to change of time, follows or may follow different laws before and after 

 the principal phase. Thus at Limerick, after low water, the water sometimes rises 

 as much in ten minutes as it had previously dropped in two hours ; it therefore appears 

 right here, if several successive observations about low water are registered at the 

 same decimal of a foot, to suppose that the real low water is little before the last of 

 those observations. At some other stations this circumstance does not happen uni- 

 formly; and then, when it does happen, it becomes difficult to say whether there is 

 a difference of law before and after the low water (in which case the real low water 

 ought to be taken nearer to the last observation), or whether the surface of the water 

 at the last observations on the same division has been depressed by accident (in which 

 case the real low water ought to be taken nearer to the first observations). 1 will 

 not undertake to say that, in marking off the times of high and low water, I have 

 followed a uniform method in these difficulties ; but I have certainly followed a uni- 

 form plan for each station ; and this is all that is important. 



Occasionally, though rarely, observations of high and low water were interrupted 

 by the roughness of the sea and other accidents. It was highly desirable to supply 

 these, because (as will be seen in the next section) differences of the heights and of 

 the times to the fourth order were to be taken, and thus the omission of one height 

 or time would entail the loss of five results in these differences of the 4th order. 

 The following is the process by which they were supplied. It very soon became 



MDCCCXLV. C 



