468 



NATURE 



[August 14, 19 19 



a source of sound has been overcome in a number 

 of ways. One type of directional hydrophone is 

 shown in Fig. i. In this instrument hoth sides 

 of the sensitive receiving diaphragm are in con- 

 tact with the sea, the microphone being encased 

 in a small capsule at the centre of the diaphragm. 

 If used in this form the instrument is deaf to 

 sounds in its equatorial plane, but can hear 

 sounds coming from other directions. It is, in 



directional hydrophone. 



fact, the reciprocal of the hypothetical "double 

 source " of Helmholtz. The polar curve, showing 

 the dependence of its response upon its orientation 

 with respect to the source, is given in Fig. 2. 



It is obvious that the ambiguity involved in the 

 bi-directional qualities of such an instrument 

 would seriously diminish its efficiency in actual 

 practice, and accordingly a modification was intro- 

 duced to eliminate this defect. This consisted in 



Fig. 2. — Direction-sensitiveness polar curve of a bi-directional hydrophone. 



the attachment to the hydrophone carcass, at 

 some distance away from the sensitive diaphragm, 

 of a bias plate, or "baffle," as it is now called. 

 This can be seen in the side view of Fig. i . 

 When correctly adjusted in position, the "baffle " 

 modifies the polar curve of Fig. 2, so that it takes 

 the form shown, in Fig. 3, and, as can be readily 

 seen, renders the hydrophone uni-directional. 

 The construction and properties of "baffles" 

 NO. 2598, VOL. 103] 



are very interesting, and have been the subject 

 of prolonged investigation. The mathematical 

 theory of their action has not been worked out 

 fully, as it is difficult to specify all boundary con- 

 ditions. Moreover, the phenomena are of the 

 diffraction type, in which the obstacle is small 

 compared with the wave-lengths of the incident 

 disturbances. A fairly complete empirical know- 

 ledge of their properties has, however, been 

 obtained. The essential feature of their construc- 

 tion is the inclusion of a film of gas in a non- 

 resonant enclosure. If the "baffle " is placed too 

 close to the receiving diaphragm, the hydrophone 

 becomes non-directional, a limiting case being 

 that in which one side of the diaphragm is com- 

 pletely enclosed, and, therefore, "over-baffled." 



In his lectures Prof. Bragg also briefly 

 described two other methods by which the direc- 

 tion of an under-water source of sound could be 

 ascertained by making use of a number of hydro- 

 phones which do not themselves possess intrinsic 

 directional properties. In the first of these use 

 is made of the binaural principle. Two hydro- 

 phones are mounted on a rotating arm at a dis- 



PUNEof Diaphra gm 

 ^ /Z>d 



Fig. 3. — Direction-sensitiveness polar curve of a 

 unidirectional hydrophone. 



tance apart of from six to eight feet, one hydro- 

 phone being connected to the right ear-piece of 

 the observer's telephone, and the other to his 

 left ear-piece. If now the wave-front of the on- 

 coming sound strikes the right-hand hydrophone 

 first, the sound appears to come from the 

 observer's right. On rotating the arm the hydro- 

 phone on the left side can be advanced so that 

 the sound appears to come from the left. By 

 rotating the device until the sound appears to 

 come from ahead or astern, the observer is 

 enabled to detect the direction of the source, a 

 simple rule enabling him to resolve any fore-and- 

 aft ambiguity. Instead of rotating the arm carry- 

 ing the hydrophones, the angle which the wave- 

 front makes with it can be found by compensating 

 for the difference of path in water by introducing 

 an equivalent length of air column between one 

 or other of the observer's ear-pieces and his ear. 

 In this case three hydrophones have to be used 

 in pairs in order to obtain the direction of the 

 source vmiquely, the bearings being read off from 

 the calibrated scale of the "compensator." 



The second method consists in making use of 



