14 A TEXTBOOK OF OCEANOGRAPHY 



irregular near the land, but also in the open océan a corre- 

 sponding irregularity is met with, sinçe each dépression of the 

 océan bottom is correlated with a dépression of the sea-level. 

 Thèse irregularities of sea-level can be determined by experi- 

 ment or can be calculated from theoretical considérations. The 

 pendulum is used for this détermination. If a pendulum be 

 swung on the same parallel of latitude over the land and then 

 over the sea where there is a marked dépression of the océan 

 floor, it should swing more quickly in the latter position, where 

 it is nearer the earth's centre. Records of the seconds pendulum 

 taken oflf the Bonin Islands (Japan) and the north coast of 

 Brazil showed, according to Listing, a différence in level and 

 consequently a height for the continental wave of 2 kilomètres. 

 This estimate is almost certainly erroneous and far too high. 



More modem attempts with Sterneck's half-second pen- 

 dulum are certainly more accurate, and the best of thèse were 

 recorded by Scott Hansen of the Fram when Nansen's ship 

 was frozen in in the Arctic (1894-95). Scott Hansen found in 

 areas of the sea where the depth was 1,640 fathoms (3,000 

 mètres) that the periodicity of Sterneck's pendulum only 

 departed very little from the normal, and it is very probable 

 that this is true, not only for the Arctic " Océan," but also for 

 the larger océans. 



Comparisons of the pressure by the mercurial barometer 

 and observations made with délicate boiling-point thermo- 

 meters on the South Atlantic between Lisbon and Bahia show 

 the gravity on this route is nearly normal. 



Temporary différences in sea-level are due to a number of 

 causes, among which may be mentioned the sun's rays; the 

 influence of large volumes of water derived from land drainage, 

 very noticeable at Kronstadt (Gulf of Finland) and Odessa ; 

 prevailing winds ; and the barometric pressure. 



Tlie level of the sea varies with the pressure of the air; 

 every movement of the mercurial barometer is correlated with 

 a change in the sea-level. Sir James Ross, who wintered at 

 Port Leopold in 1848, made hourly observations of the height 



