161 



To summarize, starting with short period waves and going toward the 

 longer periods, it appears that optical, electro-mechanical, pneumatic, numeri- 

 cal and astronomical methods seem to be well adapted to cover subsequent 

 ranges of the ocean wave spectrum. Each method extends over a range of peri- 

 ods of about 100:1. These opinions are subject to change without notice. 



A word about tides. So far we have considered their elimination by 

 suitable filters in order to study low amplitude oscillations of non-tidal origin 

 in the adjacent portion of the wave spectrum. This should not imply that tides 

 are without interest. To the contrary, recent developments make it seem quite 

 possible that with a deternnined effort very considerable progress could now be 

 achieved. In the first place, I think that an instrument can now be developed for 

 measuring tides in the open sea. In the second place, it should be possible with 

 the aid of an electronic computer, to obtain numerical solutions to the differential 

 equations of tide generation, subject to the boundary conditions as they actually 

 exist. 



DISCUSSION: Carl Eckart 



It would be redundant to go further into any of the topics that Mr. Snod- 

 grass has treated, but it occurs to me that it may be profitable to consider the 

 recording tide gauge, which is the ancestor of all wave measuring devices. In 

 age and reliability, this instrument rivals the reversing thermometer. It was 

 well-known seventy years ago. 



On March 1, 1882, Lord Kelvin read a paper entitled, "The Tide Gauge, 

 Tidal Harmonic Analyzer and Tide Predicter", before the Institution of Civil 

 Engineers.* Kelvin is perhaps most famous for his abstract formulation of the 

 second law of thermodynamics and his almost equally abstract definition of tem- 

 perature. It is perhaps not so well-known that he concerned himself with the 

 "nuts and bolts" of instrumentation. 



In reading this paper, one should remember that, seventy years ago, 

 precision machine tools were just being perfected, and that these made possible 

 the construction of many ingenious mechanisms that had hitherto been impossi- 

 ble. These ingenious mechanisms fired the popular imagination in much the 

 same way that ingenious electronic circuits do today. In fact, there were many 

 who allowed their enthusiasm for these mechanisms to influence their judgment, 

 just as enthusiasm for electronics is apt to influence judgment today. 



In particular, the fountain pen had just been invented, and Kelvin had em- 

 ployed it in his tide predictor. He had been criticized for not also using it in 

 the tide gauge: hence, the following passage from his paper. "The ink-marker 



has been tried for tide gauges both by the Author and by others, but has 



hitherto been found unsuccessful, on account of the slowness of the motion, and 

 the long time through which the action has to be continued; and as there is ample 

 driving power in the tide gauge, there is not the strong reason that there is in the 

 tide predicter for preferring the ink-marker to the pencil. " 



This passage contains what nnight be called Kelvin's First Law of Oceano- 

 graphic Instrumentation: There is ample driving power in the sea. It seems to 

 me that all successful and reliable oceanographic instruments capitalize on this. 

 The reversing thermometer has nearly one hundred times the heat capacity of a 

 clinical thermometer, and the bathythermograph probably another thousand 



* - Reprinted in Math. & Phys. Papers, Vol. VI, p. 272, Cambridge (1911). 



