10 



Fishery Board for Scotland. 



The arithmetical mean value of the discrepancy is -14 ft. at Dundee, 

 •15 ft. at Aberdeen ; and the Median value is -11 ft. at Dundee, 

 •12 ft. at Aberdeen. It will be seen that these monthly discrepancies 

 themselves fall into an orderly series, being greatest in the month of 

 February, and falling to a minimum in the month of July (Fig. 4). 

 The mean monthly discrepancy varies from year to year, as is 

 set forth in the following Table (Table D) ; and, while there is no 

 obvions regularity in the nature of this variation, it will be seen 

 (especially after" the successive annual values are smoothed, in'groups of 

 three) that at our two stations the variation is similar and concordant. 



Table D. — Mean Sea Level. Mean Monthly Discrepancies, in in- 

 dividual Years, from the Monthly Means for the entire period 

 (in fractions of a foot). 



For comparison with oiu' results at Dundee and Aberdeen, I ob- 

 tained some time ago, by the kindness of Admiral Mostyn Field, 

 F.R.S., at that time Hydrographer to the Admiralty, the tide-gauge 

 records at Milford Haven for the years 1886 to 1892. These records 

 were complete, save for a short interruption in the spring of 1891. 

 The mean sea level was determined in the same way as for Dundee, 

 that is to say, not by an integxation of the cm'ves, but by a simple 

 comparison of the high and low water levels. The results are briefly 

 set forth in Table VIII. (p. 44). 



It will be observed that the results are less regular than are those 

 for Dundee, the mean monthly values giving a more irregular curve ; 

 and this is not only due to the shorter period over which the mean 

 values are taken, but also to a greater irregularity, or a greater com- 

 plexity, of the curves for the individual years. But, nevertheless, 

 the monthly mean values show a curve which in its main features 

 agrees with the Dundee one, showing, that is to say, a spring minimum 

 and a winter maximum, with a total range or amplitude of about 

 •6 ft. And when we smooth the Milford Curve, by averaging the 



