DRS. GUY BARLOW AND H. B. KEENE ON THE ANALYSIS OF SOUND. 
143 
(5) In a similar way tlie fundamental frequencies were found for three metal dia¬ 
phragms (3 in. diameter). The values obtained were 345, 890, 1520/sec. 
(6) Highly damped vibrations. The resonator box belonging to a standard 384 fork 
was tapped in front of the trumpet of the magnetophone. The maximum deflec¬ 
tion was obtained at 344/sec., and the fork itself gave a response at 382/sec. 
In those cases in which the vibrations are not sustained there is no regular response, 
but merely a kick on the galvanometer for each excitation. At any one speed the kicks 
may be of either sign, and they vary in magnitude over a certain range according to 
the phase-difference. The frequency is determined by the interrupter speed for which 
the range is a maximum, but, as would be expected, this maximum becomes ill-defined 
when the vibrations are highly damped. 
In the above experiments electrically maintained forks were not used owing to their 
direct magnetic action on the magnetophone. 
(b) Carbon Grannie Microphone Receiver. 
An ordinary commercial instrument (G.E. Co.) of the “ solid-back “ type with conical 
mouthpiece was suspended by thin rubber cords. Some preliminary experiments in 
which the receiver was enclosed in an exhausted vessel showed that no appreciable 
vibrations were communicated through the supports to the microphone when suspended 
in this way. 
The electrical connections are shown in fig. 6. When required an additional resistance 
was placed in the galvanometer circuit to 
reduce the sensitiveness. As compared with 
the magnetophone the microphone has the 
following advantages:— 
(i.) Greater sensitiveness. 
(ii.) It is non-inductive, and therefore not 
affected by stray alternating fields 
from electric forks. &c. 
The disadvantages are :— 
(i.) Resonance : The above receiver has a 
natural frequency about 1030/sec. 
(ii.) The mean resistance during vibration differs from the normal resistance (it is 
actually increased), and therefore the starting and stopping of a sustained 
sound gives kicks (in opposite directions) on the galvanometer. 
(iii.) The resistance, and consequently the sensitiveness, is therefore subject to uncon¬ 
trollable variation. 
(1) Pure Tones. —A fork 512, mounted on its resonator, when bowed or struck near 
the receiver gave, on analysis, deflections off the scale. 
B/vcxl Galvcuzointber 
-nsm- 
90 co 
^ Interrupter 
1 MMiarruneta’ 
r-iMt 
T Volts 
Microphone 
Fig. 6. 
