6 Trans. Acad. Sci. of St. Louis. 



behavior of the mica windmill as described in the former 

 paper, p. 67. 



Some work has been done in the study of the effect 

 of imposing sonnd vibrations upon a column of air in a 

 resonance tube, along which an electric discharge is pass- 

 ing. , It has been found that the electrical discharge is 

 affected by the system of standing waves. A luminous 

 column has thus been converted into a dark discharge. 

 The musical tones which accompany the surging to and 

 fro of the air molecules in a gap, have been somewhat 

 reinforced by placing the discharge gap within a reso- 

 nance tube, whose length could be adjusted to the musical 

 note. 



The air vibrations which are thus produced are, how- 

 ever, not in harmony with the conditions which must 

 exist in a system of standing waves. A complete wave 

 consists of a dark space and an adjacent luminous col- 

 umn. The air molecules in these semi-waves are being 

 urged in opposite directions. The nodes where at any 

 instant we may suppose conditions of alternate maximum 

 and minimum pressure to exist, along the discharge, lie 

 between the dark and the luminous halves of the waves. 

 At the same time these nodes have been transformed 

 into points of maximum and minimum conduction. These 

 are, however, the conditions which exist midway between 

 nodes. Such electrically produced sound waves will thus 

 have a progressive motion. This progressive motion is 

 maintained by the forced transfer of the negative cor- 

 puscles, which is in the same direction in the conduction 

 and in the convection halves of each wave. 



The prolonged roar which accompanies a lightning dis- 

 charge may be brought about by such a system of electri- 

 cally produced sound waves. A rattling vibration may 

 often be heard in overhead thunder, which strongly sug- 

 gests ripples superposed on a conduction discharge. Of 

 course reflections from distant cloud-like masses of air, 

 and effects due to varying distances of the discharge 

 channel from the observer are not here referred to. 



