228 DR. LOUIS YESSOT KING ON THE PROPAGATION OF SOUND IN THE FREE 
generators, the estimates of relative intensity being made by ear and indicated by 
numbers between 1 and 10. For long-range work the superiority of the motor-driven 
siren operating at about 25 pounds air-pressure was definitely established ; in weather 
conditions associated with a smooth sea, a pitch of 182 complete vibrations per second 
was found to be most suitable, while a high pitch note of 295 vibrations per second 
was heard to better advantage during high wind and a rough, noisy sea. 
The scientific aspect of the preceding tests is discussed at some length by Lord 
Rayleigh in a later paper, # it is there pointed out that the increase of power required 
to operate the sirens so as to generate more powerful sound-waves is entirely out of 
proportion to the gain in range of audibility. This point is well illustrated by the 
fact that when the high note of the Scotch siren was sounded, energy was consumed 
at the rate of about 600 horse-power. The experiments just referred to constitute 
a final practical test of particular forms of sirens, in that the signals are judged under 
the same conditions and by the same means as when they are issued as signals to 
navigators. While there is no difficulty in thus estimating the relative merits of 
various forms of sound-generating apparatus it is evident that very little information 
is to be derived in this way as to the proportion of power actually converted into 
sound or as to the causes of the very large waste of energy which is known to occur 
in all sound-producing instruments. 
An account of recent researches carried out in France is given very briefly by 
Ribiere, | who mentions several unsuccessful attempts to construct instruments for 
the purpose of measuring sound-intensity. In the French Lighthouse Service a study 
of the influence of atmospheric conditions has been made statistically by keeping 
daily records at various fog-signal stations of the audibility of daily test-signals sent 
out from neighbouring stations. Ribiere also points out that the sound-waves 
emitted by a powerful siren are probably different in character from waves of 
ordinary intensity, and that in the former case the differential equations for waves 
of small amplitude no longer apply. The question of the acoustic efficiency of fog- 
sirens is mentioned as one of the unsolved problems of the subject and the importance 
of its solution is emphasized. 
In order to obtain experimental data to form a basis for a discussion of the various 
problems mentioned above, it was thought desirable by the writer to undertake 
a systematic investigation of the subject with the experimental means now available 
for attacking 1 the difficulties mentioned. Owing to the elegant construction of the 
form of siren adopted in recent years by the Canadian Government (the “ diaphone”) 
and the ease with which the air pressure and air consumption could be controlled and 
varied, successful efficiency tests were carried out by the thermodynamical method 
* Rayleigh, “On the Production and Distribution of Sound,” ‘Phil. Mag.,’ vol. 6, pp. 289-305, 1903. 
‘ Scientific Papers,’ vol. v., No. 126. 
t Ribiere, C., ‘ Phares et Signaux Maritimes,’ Octave Doin et Fils, Paris, 1908. In this work references 
are given to important technical papers on the subject of fog-signals up to the year 1908. 
