13 



II. TIDAL OBSERVATIONS OF THE "SCOTIA," 1902-1904. 



I. ANALYSIS OF THE OBSERVATIONS. 



The anchorage at which the tidal observations here dealt with were taken was at the head of Scotia 

 Bay, Laurie Island, South Orkneys, in latitude 60 43' 42" S., longitude 44 38' 33" W. The bay is about 

 3 miles deep and faces S.E. ; the depth of water at the anchorage was 10 fathoms, increasing to about 100 

 fathoms at the mouth of the bay. Here the "Scotia" was frozen in from March 25 to November L'.'i, 1903. 

 (See Map at end of volume.) 



The apparatus used consisted of a long wire fixed to the sea bottom by means of a heavy weight. The 

 wire was carried over a davit by means of a pulley. At the extremity of the wire was another lighter 

 weight, which rose and fell with the tide along a graduated wooden scale. The floe in which the 

 "Scotia" was frozen moved with the tide, the height of which was thus shown by the position of the 

 movable weight. The observations were made by the leader, officers, and staff of the Expedition under 

 the direction of Captain THOMAS ROBERTSON. 



The period covered by the observations is from 10.30 a.m., May 26, 1903, to midnight on October 16 

 of the same year. On September 4 the wire broke and had to be replaced, and at this point there may 

 be a discontinuity of zero. The observations for the last month have accordingly been treated as a 

 distinct series. The observations recorded give the height of the tide at every half-hour from May 26 to 

 September 4, and at every hour from September 5 to October 16. Eecords of the barometric pressure and 

 of the strength and direction of the wind are also available. On May 26 the height of the tide was noted 

 every 5 minutes from 5.45 a.m. to 9 a.m. 



In analysing the observations hourly heights were used, commencing for the first three months from 

 1 a.m. on May 26, 1903 (13 h. May 25, local mean time). The harmonic components for which an 

 analysis was carried out were M->, 0, N 2 , 82, K 2 , KI, P. 



For the first three components named tho method of analysis was that used for the Indian Tides, and 

 described in the 'B.A. Report' for 1883. The periods chosen were 86 M days, 89 days and 78 N days, 

 these periods being selected to minimise the effects of the S_., K,, and M 2 components respectively, and 

 to be as large as the first series of observations permitted. 



For the components S-> and K- 2 and KI and P the process employed was essentially that given by 

 (Sir) G. H. DARWIN in the 'R.S. Proceedings,' vol. LII (1S9.'5), p. 365, win-re a iin-tlnxl is indicated for 

 dealing with a short series of observations extending over a few months. In analysing for KI and P, 

 however, 27 days' observations in each .month were employed instead of 30. [See Article by (Sir) 

 G. H. DARWIN in Appendix to 'B.A. Report' for 1886.] Each month's observations were separately 

 analysed and the means formed as recommended on p. 367 of the paper referred to above. 



The most important feature of the method, in regard to the results of the analysis tabulated below, is 

 that the ratios of the amplitudes of 83 and K>, and of KI and P respectively are assumed to have their 

 theoretical values, and that the " lags," in the case of each pair of components having nearly equal speeds, 

 are taken to be the same. That in this instance this theoretical assumption is in sufficiently close 

 agreement with the facts may be regarded as established by the results of the analysis. 



The values of the tidal constants obtained from the analysis are given in the subjoined table, H denoting 

 the semi-range or amplitude (in feet) and K the epoch or " lag " of a component : 



