axi internal wave along this boundary, with a resultant hind- 

 rance of the forward motion of the ship. 



Vertical displacements of water masses may be easily 

 observed by repeating, at short time intervals, oceanographic 

 observations at one place. The first investigation of this 

 type was made by Helland-Hansen and Nansen (1909) on the 

 Michael Sars expedition. During the cruise repeated serial 

 observations were made by two ships and a harmonic analysis 

 was made of the data. 



If such observations are repeated for a long enough time, 

 it is possible to find by harmonic analysis the period of the 

 waves. Although short periodic internal waves such as were 

 described by Ekman are known to exist and are actually observed, 

 the Michael Sars data yielded the information that the predom- 

 inant periods of the oscillations were of semidiurnal and di- 

 urnal nature. 



This conclusion has been borne out by the work of other 

 authors. One of the most complete analyses of such data was 

 made by Selwell (1939, 19l|-2). The results of this investigation 

 showed that, after the oscillations of 12- and 2l|--hour periods 

 were eliminated, the residue exhibited a normal or random dis- 

 tribution, Seiwell also demonstrated that, not only did the 

 diurnal and semidiurnal v/aves predominate in period, but they 

 also made the greatest contribution to the amplitude of the 

 viaves , 



Although these Investigations have demonstrated that the 

 vertical oscillations in the oceans are dominated by waves of 



34 



