286 Lieut. -Colonel C. B. Heald and Major W. S. Tucker. 



The air, with its small inertia, here takes the place of the mechanism of the 

 swinging table. The movements of the air are recorded faithfully and 

 without measurable lag by the change in the temperature and electrical 

 resistance of the hot-wire microphone. 



The microphone is subject to resistance variation through change in the 

 temperature of the room, thus changing the zero. This effect, however, can 

 be overcome by simply adjusting the balance of the Wheatstone bridge. 



In order to use the microphone — which has a resistance of about 150 ohm 

 when heated — it is inserted in one arm of a Wheatstone bridge, the other 

 arm being adjusted by a rheostat to give balance with the string galvano- 

 meter. The pulsating currents of air which cool the microphone grid, 

 give a corresponding variation of resistance, but it must be noted in this 

 investigation that motions of the air, whether positive or negative in direc- 

 tion, always produce, positive deflections on the galvanometer wire, since all 

 air movements produce a fall in temperature and a diminution in resistance 

 of the microphone grid. 



Another point to be emphasised is that deflections of the galvanometer 

 wire are not proportional to the displacements of the body in a given direc- 

 tion, but are proportional to some function of the velocity of the body under 

 recoil. What this function is, may be indicated by reference to the previous 

 work done on the theory of the hot-wire microphone. 



The deflection of the galvanometer, which may be considered to be pro- 

 portional to the change in electrical resistance of the microphone, is 

 dependent not only on the vibratory motions of the air set up by the heart 

 •action, but also on a certain amount of direct air current set up by convec- 

 tion, as a result of the disposition of the microphone grid in the orifice 

 through which the vibrations are transmitted. Convection effect was 

 reduced to a minimum in this investigation, by setting the plane of the 

 microphone grid vertical, so that convection currents tend to be perpen- 

 dicular to the displacements of the vibrating air. These currents, however, 

 could not be completely eliminated, owing to the lack of complete symmetry 

 of the enclosure on the opposite sides of the grid. 



It has been shown in the paper above referred to that for a certain type of 

 grid similar to that in use and under similar conditions, receiving vibrations 

 U sin pt, the total resistance change 



SR = - 0-15 U 2 + 015 Vsmpt+ 015 U 2 cos 2pt, 



where U is the velocity of the air at any moment past the grid, 2tt /p the 

 period of vibration of the air. Other terms may be added to the above 

 series, but have been shown to be so small as to be negligible. The con- 



