SECT. 5] LONG-TERM VARIATIONS IN SEA-LEVEL 593 



reference level and Aa the anomaly in specific volnme. Despite uncertainties in 

 determining the reference level, such investigations are highly profitable, 

 particularly when they are conceived and carried out on a grand scale. Indeed, 

 it is highly doubtful whether such an investigation would be successful in 

 representing steric sea-level variations at one station by means of locally 

 observed density unless the steric variations were of considerable magnitude. 

 This follows from the works of Pattullo et al. (1955) for a world-wide network 

 of stations, and of Lisitzin, Polli, Hela and others for more limited regions; 

 these clearly show that seasonal variations in steric sea-level can be considered 

 regional phenomena, varying little over quite wide areas (see Pattullo in Vol. 

 2). In this context a pertinent question is whether a thin layer of surface water 

 suffering large density changes is more efficacious in producing changes in 

 steric level than either a thick layer of deeper water suffering smaller density 

 changes or the seasonal incursions of bottom water into lower latitudes. The 

 answer almost certainly depends on the particular area under consideration. 

 The English Channel and the North Sea are rather shallow areas of strong tidal 

 mixing ; it might be hoped, therefore, that surface densities would reflect quite 

 closely the density structure in depth in so far as annual variations are con- 

 cerned. Contemporary investigations by the writer for stations in these areas, 

 using 19 years of observations, show no correlation between surface layer 

 densities and observed annual means of sea-level. This may mean that in these 

 regions either the properties of the surface layers are not paramount in deter- 

 mining steric levels, or that year-to-year variations in steric level are 

 insignificant. 



One further direct oceanographical factor affecting sea-level in high latitudes 

 is the damping influence sea-ice has upon the vertical movement of the water 

 surface. No estimate appears to have been made of its magnitude for long-term 

 variations, though it is likely to be rather small. 



B. Climatological Factors 



The last section indicated the indirect effect of a meteorological agency (the 

 wind) in changing sea-level through the mechanism of steric changes. The more 

 direct effects on sea-level of atmospheric pressure and the tractive force of the 

 wind have long been studied. The statical law of barometric influence in which 

 the sea acts as an inverted barometer, rising approximately 1 cm for a decrease 

 in air pressure of 1 mb, and vice versa, was first examined by Gissler in 1747 

 for the Gulf of Bothnia, and since by many others for other regions. The 

 greatest difficulty experienced in verifying the law has arisen from the presence 

 of the much larger effect due to the tractive force of the wind, and whilst the 

 theoretical law is invariably assumed as a sufficiently close approximation to 

 the truth, a conclusive confirmation is still lacking. The outstanding point at 

 issue concerns the minimum period of time necessary for the inertia of the 

 water to be overcome in order that the statical law be obeyed. Doodson (1924) 

 has shown that for day-to-day variations in sea-level the theoretical statical 

 coefficient is approached, for British waters, to within 15%. 



20— s. I. 



