496 



SCIENCE. 



[Vol. VII., No. 174 



being placed in communication with a reservoir 

 of air under gentle pressure (fig. 1). Vibratory 

 motions being then excited in the diaphragm, by 

 means of a battery and a microphone or rheotome 

 in a distant apartment, the discovery was made 

 that speech as well as musical and other sounds 

 could be quite loudly reproduced from the flame. 

 Certain observations led the author to suspect that 

 motion of the orifice, rather than compression of 

 the air in the chamber, was the chief agent in the 

 phenomenon ; and, in fact, precisely similar re- 

 sults were obtained when a light glass jet-tube 

 was cemented to a soft iron armature, mounted 

 on a spring in front of the telephone magnet 

 (fig. 2). _ 



Experiment also showed that an air-jet at suita- 

 ble pressure directed against a flame repeats all 

 sounds or words uttered in the neighborhood (fig. 

 3). Except, however, where the impressed vibra- 

 tions do not differ widely in pitch from the nor- 

 mal vibrations of the jet (discovered by Sondhauss 

 and Masson), these effects are likely to escape 

 notice owing to the inability of the ear to dis- 

 tinguish between the disturbing sounds and their 

 echo-like reproduction from the flame. 



In these experiments the primary action of the 

 impressed vibrations was undoubtedly exerted on 

 the air- jet ; but a singular and perplexing fact 

 was that no sound, or at best very faint sounds, 

 could be heard from the latter when the flame 

 was removed, and the ear, or the end of a wide 

 tube connected with the ear, was substituted for 

 it. Suspecting, finally, that the changes in the jet, 

 effective in producing sound from the flame, must 

 be relative changes of different parts of it, the 

 author was led to try a very small hearing-orifice, 

 about as large as the jet-orifice (fig. 4). The re- 

 sults were most striking. By introducing this lit- 

 tle hearing-orifice into the path of a vibrating 

 air- jet, the vibrations can be heard over a very 

 wide area. Close to the jet-orifice they are so faint 

 as to be scarcely audible ; but they increase in 

 intensity in a remarkable way as the hearing- 

 orifice is moved away along the axis of the jet, 

 and reach their maximum at a certain distance. 

 Experiments with smoked air showed that this 

 point of maximum sound is that at which the jet 

 loses its rod-like character, and expands rapidly : 

 it has been named the ' breaking-point,' because 

 just beyond it the sounds heard from the jet 

 acquire a broken or rattling character, and at a 

 greater distance are completely lost. The distance 

 of the breaking-point from the orifice diminishes 

 as the intensity of the disturbing vibrations is in- 

 creased, and also depends to some extent on their 

 pitch and on the velocity of the jet. With orifices 

 of from 1 to 1.5 mm. in diameter, it usually varies 



from 1 to 6 cm. The vibrations of an air-jet may 

 also be heard at points not situated on the axis ; 

 but they are always most intense along the axis, 

 and become rapidly fainter as the distance from it 

 increases. 



With glass jet and hearing-tubes, and a light 

 gas bag to serve as reservoir, these experiments 

 are easily repeated ; but simple apparatus for 

 more careful experiments is described. The au- 

 thor's general conclusions from his experiments 

 and those of others are as follows : — 



A jet of air at moderate pressure (below 10 

 mm. of water) from an orifice from 1 to 1.5 mm. 

 in diameter, forms a continuous column for a cer- 

 tain distance, beyond which it expands and be- 

 comes confused. 



Any impulse, such as a tap on the jet support, 

 or a short and sharp sound, causes a minute dis- 

 turbance to start from the orifice. This disturb- 

 ance increases in area as it progresses, and finally 

 causes the jet to break. By directing the jet 

 against a flame or a hearing-orifice, it is readily 

 perceived that such disturbances travel along the 

 jet-path with a velocity which is not that of sound 

 in air. In fact, the sound heard in the ear-piece 

 resembles an echo of the disturbing sound. 



The disturbances produced by sounds of differ- 

 ent pitch travel along the jet-path with the same 

 velocity. This is evident, since otherwise ac- 

 curate reproduction of the complex vibrations of 

 speech at a distance from the orifice would be 

 impossible. This velocity is much less than that 

 of sound in air, and is probably the mean velocity 

 of the stream. 



A vibrating air-jet playing into free air gives 

 rise to very feeble sounds, but these sounds are 

 much intensified when the jet impinges on any 

 obstacle which serves to divide it into two parts. 

 Of such arrangements, the best is a perforated 

 surface, the orifice being placed in the axis of the 

 jet. 



A jet of air at low pressure responds to and 

 reproduces only sounds of low pitch. Sounds 

 above a certain pitch, which depends on the press- 

 ure, either do not affect it or are only faintly 

 reproduced. 



At pressures between 10 and 12 mm. of water, 

 an air-jet reproduces all the tones of the speaking 

 voice, and those usually employed in music, with 

 the exception of very shrill or hissing noises. 

 When the pressure in the reservoir equals about 

 13 mm. of water, hissing sounds are well repro- 

 duced, while sounds of low pitch become fainter. 

 At higher pressures, up to about 25 mm. of water, 

 shrill or hissing noises produce very violent dis- 

 turbance, while ordinary speech tones have little 

 effect. But at these pressures sounds of high 



