394 



SCIENCE 



[N. S. Vol. LIII. No. 1373 



air grows gradually less until it becomes zero 

 at the node N. Compressions and rarefac- 

 tions have maximum values at N and grow 

 less in intensity until at A there is no change 

 in pressure. The oscillating motion will per- 

 sist for a number of vibrations until the fric- 

 tion with the sides of the tube gradually 

 brings about a condition of rest. If the 

 motion is to be maintained, energy must be 

 supplied to the vibrating system. In Dr. 

 Knipp's apparatus, this energy is furnished at 

 cc by a ring burner gas flame. 



As the air surges up the tube and turns to 

 pass into D, it is brought into intimate con- 

 tact with hot glass at CC. This portion of 

 heated air communicates its energy to sur- 

 rounding air particles at D and results in a 

 general rise of temperature in this region. 

 It tends to expand the air. Just at this time, 

 the air begins to surge in the opposite direc- 

 tion throughout the tube and the tendency to 

 expand at D assists this return surge. 



When the air surges back from D to EE it 

 becomes heated again at C and the heat ac- 

 quired is given up partly to the glass at EE, 

 thus causing a cooling and consequently con- 

 traction. In the meantime, the air has 

 finished its outward motion and is ready to 



r~^ 



Fig. 1. 



surge back toward the inner part of the tube. 

 The cooled air at EE by its contraction assists 

 this motion. Thus, the motion is maintained 



first by the heating effect (expansion) at D 

 and half a period later by the cooling effect 

 (contraction) at EE. 



The alternating motions are quite rapid 

 since the vibrations in the various types of 

 tubes range from 30 to 250 times per second. 

 It is well established that air may be heated 

 and cooled thus rapidly because the accepted 

 explanation for the propagation of sound in 

 air requires this rapid change in temperature. 



The amplitude of tlie motion at CC is not 

 great but appears to have a range from 0.5 to 

 1.5 cm., depending on the tube used. The 

 lower pitched tubes have the larger amplitude. 

 The area of glass heated is small so that the 

 temperature gradient from the red hot glass 

 to the cooler portion 0.5 cm. distant may be 

 several hundred degrees, thus allowing a con- 

 firmation of the theory advanced. 



The additions and subtractions of heat to 

 the air are not impulsive but, by the processes 

 of transmission explained, gradually build up 

 to maxima at suitable instants to maintain 

 the motion. Cooling the tube at BE assists 

 the vibration. The motion is maintained 

 more easily if the area of the annual ring at 

 EEC is made smaller than the area of the 

 tube at D. This results in a greater ampli- 

 tude of motion at C in agreement with the 

 theory. 



Mechanically, the motion may be likened to 

 the motion of a system consisting of a spring 

 that is fastened at one end with a weight at 

 the other end that vibrates back and forth. 

 (See Fig. 2.) When the mass m moves to the 



Pig. 2. 



left, the spring is compressed and presently 

 when equilibrium is reached, the motion 

 reverses. If, at tliis instant, the spring at D 

 could be strengthened by the insertion of 

 several coils, the return motion would be 

 assisted. This corresponds to the addition of 

 heat in Dr. Knipp's tube. 



Wlien the spring has expanded to the other 

 limit of its vibration, imagine several loops 



