A SELECTIVE HOT-WIRE MICROPHONE. 
395 
The grid carries, normally, a heating current of about 27 milliamperes. When the 
resonator responds to a sound, the to-and-fro motion of the air in the neck produces, 
as already stated, an oscillatory change of resistance, and the effect of this is to super¬ 
impose on the steady heating current a ripple of small amplitude (generally only a few 
microamperes). It is this ripple which is amplified and which is heard in the telephones. 
It is shown in a later paragraph (§7) that the magnitude of the amplified current 
may be used as a means of estimating the amplitude of a sound. For this purpose 
the telephones are replaced by a vibration galvanometer tuned to the pitch of the sound. 
This method of employing the microphone for the measurement of a sound, however, 
is not altogether satisfactory, on account of the difficulty of maintaining an amplifier in 
such a condition that the current amplification is constant for any length of time; and 
for this reason the Wheatstone’s Bridge method is sometimes preferable. The 
advantages of the Amplifier Method are that it is very sensitive (especially when a 
vibration galvanometer is used) and that the microphone can be placed in a moving 
piece of apparatus, subject only to the restriction that its axis must always be vertical 
(or at some fixed angle to the horizontal). Vibrations communicated through the 
mounting of the microphone (even when they are produced by striking the container) 
have very little effect on the sound heard in the telephones. 
(ii) The Wheatstone’’s Bridge Method .—This method is preferable to the Amplifier 
Method on account of its greater simplicity, and because there is no danger of the 
sensitivity changing during the course of a long series of observations. A convenient 
form for the Bridge to take is shown diagrammatically in fig. 4. The microphone (M) 
with the milliammeter (A) forms one arm of the Bridge. The balancing arm (R) is 
made about equal to the resistance of the grid when carrying its working current, i.e., 
about 350 ohms. The rheostat (Rh) is inserted (as shown) in series with the battery, a 
balance being obtained by adjusting (Rh) until the current through the microphone 
brings its resistance (together with that of the milliammeter) up to R. For some purposes 
it is convenient to have a small variable resistance p in series with the microphone. In 
most experiments it is sufficient to take the deflection of the galvanometer as a measure 
of the intensity of the sound affecting the microphone ; but other methods can of course 
