Maintenance of Forced Oscillations of a New Type. 513 



For soda glass, the kind used by Professor More in his 

 electrostriction experiments, we find 



c. 



e/e'. 



+ 1-06 



0-03 



-1-06 



0-11 







0-07 



Thus the elongation to be looked for is certainly less than 

 11 per cent, of that which Prof. More expected, and his 

 conclusion that the elongations he obtained were caused by 

 heating is confirmed by these results. 



With the Jena glass tube considerably greater elongations 

 can be looked for, as shown by the following table : — 



c. e/e'. 



+ 0*62 0-22 



-0-62 0-28 



0*25 



It is jnst possible that using glass tubes of this kind a true 

 elongation caused by electrostriction might be observed. 



In conclusion, it appears that one will not go far wrong- 

 in taking for solid dielectrics, as for fluid dielectrics, g : = ^ 2 . 

 Assuming this, the calculation of ^ 2 — °"(^i + ^ 2 ) given in the 

 previous paper from Wiillner and Wien's experiments gives 

 at once the values of Si = S 2 f° r the kinds of glass which they 

 used. 



Palmer Physical Laboratory, 

 Princeton, New Jersey. 



XLVIII. 'lhe Maintenance of Forced Oscillations of a New 

 Type. By C. V. Raman, M.A* 



[Plates X.-XIL] 



^pHE photographs published with this paper (Pis. X.-XIL) 

 JL represent some of the very remarkable class of main- 

 tained oscillations regarding which preliminary communica- 

 tions were published in the issues of ' Nature ' dated the 

 9th Dec, 1909, and the 10th Feb., 1910, under the title of 

 the present paper. I propose here to confine myself to 

 indicating the main features of the phenomena actually 

 observed in experiment, and to elucidating these by the aid 

 of a physical theory that I have since worked out. 



One of the most fascinating problems in the dynamics of 

 * Communicated by the Author. 



