curve V'jlth periods of approximately one inonth, Fourier aiialydls 

 was applied to these data in an attempb to discern a true 29-day 

 cycle; the results were nen;ative. This fact is not surprising, 

 since so many other parameters affect the depth of the mixed 

 layer that any monthly variation must bo treated not as a total 

 cause in itself, but rather as a partial cause together with 

 other causes. This technique will be discussed later. 



c. Semidiurnal and diurnal tidal variations - It has already- 

 been mentioned that previous investigators have concluded that the 

 greatest short period variations In the depth of the therniocllne 

 have their origins In an internal wave of tidal character. For 

 this reason the main emphasis has been placed on attempting to 

 discover techniques which would predict this effect. 



The simplest technique one can use in this case is a purely 

 statistical one, such as is used in predicting the tides them- 

 selves. It was noticed that some of the time-series traces 

 of the thermocline variations at ship stations in the Jiiastern 

 Pacific bore a marked resemblance to the tidal curves at stations 

 on the West Coast of the United States. borne of the best data, 

 therefore, were c^iven to the U. S. Coast and Geodetic Survey 

 with a view to having the data analyzed for the presence of 

 semidiurnal and diurn'al constituents and eventual prediction of 

 the depth of the thermocline by means of constants. It was 

 fully realized that the depth and intensity of the therm-ocline 

 have an effect on both the amplit'ide and phase of the wave, but 

 it vjas hoped tbnt the results v:oald be useful for a particular 

 place by usbig a different set of ccnstanns for e?-oh location. 



43 



