time by superposition, the calibration constants would be considerably modi- 

 fied due to the fact that the depth of the submerged tanks is less. This would 

 cause the wave pole to move up in the crest of the shorter period wave even 

 more than the theory would predict. 



This effect is not compensated for by an equal and opposite effect when 

 the trough of a short period wave is present on the crest of a long period wave 

 due to the exponential behavior of these factors. Thus the response may be 

 non-linear and the heights of the shorter period -waves may be underestimated. 

 For very short period waves such effects would again be negligible, 



If the calibration of the wave pole fails to explain the discrei>ancy be* 

 tween the two sets of observations then the other possible ejqjlanations will 

 have to be investigated. On the basis of the above considerations » a predic- 

 tion is ventured that the wave pole calibration will explain the discrepancy. 



If the above hypothesis is a correct one, then the study of ocean waves 

 is in a very odd position. The w^ave pole data were to have heen a primary 

 calibration for the stereo data. The stereo data appear to have detected, to 

 the contrary, a faulty theoretical calibration of the wave pole. The shipborne 

 wave recorder developed by Tucker [1956 a] has been compared with the 

 WHOI wave pole, and agreement was not obtained in this comparison either 

 (Tucker [1956b]). This does not necessarily lead to the conclusion that the 

 shipborne instrument is correctly calibrated. In fact, its response at high 

 frequencies is known to be poor (Tucker [1956 a]), Therefore at present, 



there is no primary instrixment capable of measuring waves as a function 



of time at a fixed point in deep water. 



215 



