109 



any analysis equipment that is used after the recorder. It is very important to 

 understand the response of the whole system to the complex sound at very short 

 distances in order to understand completely the effect of the medium in altering 

 the sound both in intensity and in character. A second equally important prob- 

 lem of the same nature is to know the radiation characteristics of the sound 

 source. Is it non-directional at all frequencies? If not, what are its direction- 

 al properties? How stable is it? Does it give out sound of the same composi- 

 tion independently of time or of wear on the mechanical parts? Is it subject to 

 breakdowns for which it should be monitored? The facts about explosive sound 

 sources are interesting here. 



The selection of the explosive is determined by available supply to a 

 large extent but where possible pentolite should be used for smaller charges, 

 say, of the order of a pound. This is because pentolite requires no booster and 

 hence tends to be more reproducible. For larger charges, boostered TNT is 

 entirely satisfactory. In fact the reproducibility requirements for acoustics 

 are satisfied in most work by boostered TNT, even for small charges, since the 

 lack of reproducibility is of the order of 2 to 5 per cent. 



The shape of the charge is not especially critical where spherical waves 

 are desired. Squat cylinders or spherical charges are accurately non-direc- 

 tional, but even such cylindrical shapes as a Mk 4 demolotion block (^ pound) or 

 a conventional depth charge (300 pounds) are less directional than most so- 

 called non- directional acoustic transducers. 



Although the initial shock wave within a range of about 5 miles has a 

 simple, wide Fourier spectrum it is followed by bubble pulses with a Fourier 

 spectrum not expressible in simple analytical form. 



The use of such complex sounds introduce transmission complications 

 which must be considered in designing the experiment: the presence of reflect- 

 ing surfaces causes reflections which make the actual Fourier spectrum at the 

 point of observation vary as a function of range, charge depth and receiver 

 depth, and in the case of echo experiments with the position of the echo target 

 as well; also transnnission characteristics vary as a function of frequency, and 

 where these variations are not the object of study due account must be taken of 

 themi. 



Mechanical sources are subject to wear and tear, but the few I am some- 

 what familiar with seem to fail acoustically, suddenly and in a manner that can 

 be discerned by an experienced, alert observer. Electronically powered trans- 

 ducers can frequently be monitored acoustically. If this is not feasible, elec- 

 trical monitoring of the power to the transducer is the best substitute. 



ECHO SOUNDING AND ECHO RANGING 



Both echo sounding and echo ranging employ instrumentation of a special- 

 ized sort to send out a sound signal which travels through the water to be reflect- 

 ed from the object under study and returned to a receiver. In echo sounding 

 the usual object of study is the shape and depth of the bottom of the ocean. The 

 echo sounder is also used for studying the vertical distribution of marine ani- 

 mals through a study of the Deep Scattering Layer. We now know that the Deep 

 Scattering Layer is far more complicated than is innplied by its simple sounding 

 title. This layer may consist of one or several concentrations of sound scatters 

 at various depths. The indication on an ordinary echo sounder of the scattering 

 layer appears as reverberation, and the techniques for observing and analyzing 

 the scattering layer records are somewhat sinnilar to those developed for study- 

 ing reverberation. As is well known by now, echo ranging was developed for 



