152 
DRS. GUY BARLOW AND H. B. KEENE ON THE ANALYSIS OF SOUND. 
Sounders :— 
(i.) A platinum contact was attached to the diaphragm, and the sounder made self¬ 
driving after the manner of an electrically maintained fork. The frequency was 
then determined either by analysis of the current induced in a single turn of 
wire acted on inductively by the circuit, or by analysis of the current generated 
by a U-magnet attached to the sounder diaphragm. The platinum contact 
worked quite well under water, but this system of driving the sounder was not 
suitable for general use. 
(ii.) With the sounder immersed, the diaphragm was gently tapped with a rubber 
hammer at intervals of 1 second, and an analysis made of the current induced 
in the windings of the electromagnet, as in earlier experiments on the analysis 
of damped vibrations. 
Receivers :— 
(i.) By experiments in which a sounder was driven at various frequencies and the 
receiver used to obtain a resonance curve. 
(ii.) The natural frequency of the diaphragm was excited by impulses such as an oar- 
splash or a tap on a neighbouring boat. In the latter case the analysis showed 
in addition the natural vibration of the boat. 
(iii.) A method more satisfactory than the last was to analyse the general water 
disturbance on a rough day. (Compare experiments on Analysis of Background, 
p. 146.) 
Metal Diaphragm Receiver (Natural frequency, 120/sec.).—This instrument in mode¬ 
rately calm weather could be used, when hung from an anchored boat with a galvano¬ 
meter disturbance of no more than about 1 div., corresponding to a faint background 
in the telephone. Under rougher conditions a strong diaphragm noise was produced 
in which on certain occasions the separate impulses could be identified with the lapping 
of the water against the boat. The sounder could be distinctly heard at a distance 
of 60 feet. 
With this instrument remarkable variations of the intensity of the sound with depth 
were observed, showing the existence of an almost silent layer on the bottom. Slowly 
raising the receiver from the bottom to the surface showed a rapid increase in the first 
6 feet, followed by a slower falling off towards the surface, the maximum occurring at 
6-9 feet from the bottom. The observed amplitude at 6 feet was in some cases 25 
times that at 1 foot from the bottom. The sound in the telephone varied in a similar 
way. These effects as tested by the telephone appeared to be the same in different parts 
of the reservoir, where the depths were 12, 18 and 32 feet, and independent of the fre¬ 
quency which ranged from 60-130/sec. Sounds due to taps on the boat, splashing of 
oars, &c., and also natural disturbances, were modified in the same way. No definite 
effect of this kind was observed when using the other receivers of lower frequency. 
The amplitude of the vibration fell off with horizontal distance from the source with 
