518 Mr. C. V. Raman on the Maintenance of 



pass from without into the system to counteract the effect of 

 dissipation and thus maintain the motion. 



The class of maintained vibrations discussed above can be 

 realized in experiment and caused to yield some very beautiful 

 results that will now be described. The vibrating systems 

 used in the work are stretched strings which seem prima 

 facie well adapted for acoustical work on this subject. Im- 

 portant advantages are that their vibrations are visible with- 

 out any aid, and, secondly, that their frequency can be 

 adjusted with ease and accuracy over a comparatively wide 

 range. With stretched strings again, it is comparatively an 

 easy matter to impose a periodic variation of " spring ", i.e. of 

 tension in this case, of any desired magnitude and frequency. 

 The experimental arrangements required are fairly simple. 

 A fine string one or two metres in length is taken and one 

 end of it is passed over a peg or a pulley, preferably the 

 former. The other end is attached to the prong of a massive 

 tuning-fork of frequency say 60 or 100 per second, in such 

 a manner that the tension of the string is varied by the motion 

 of the prong. The tuning-fork should be maintained elec- 

 trically, but for some purposes powerful bowing is sufficient. 

 It is generally supposed that under these circumstances 

 the equilibrium position of the string becomes unstable and 

 it is thrown into a state of permanent and vigorous vibration 

 only when the frequency of the free oscillations of the string 

 (in its fundamental or other mode) is approximately half of 

 that of the point of attachment. I found that this was not 

 the case. A permanent and vigorous vibration was also 

 maintained when the frequency of the free oscillations of 

 the string was equal to, or one and a half times, or 

 twice, or two and a half times, or thrice that of the 

 fork. This is precisely what is to be expected from the 

 theory dealt with above, and there is absolutely no diffi- 

 culty in obtaining any one of the first five or six of these 

 types of maintained oscillation each separately by itself, 

 i. e. with the frequency of the oscillation of the string- 

 having the specified relation to that of the tuning-fork. 

 All that is required is that the tension should be so adjusted 

 till the frequency of the free oscillations of the string in its 

 fundamental mode bears the desired relation to the frequency 

 of the fork. This can be done by trial, and it will then 

 generally be found that the oscillations are maintained. 



To obtain the results shown in the photographs (Pis. X.-XII.) 

 a somewhat different procedure is adopted. The vibrating 

 system here dealt with, i.e. a stretched string, has not merely 

 one free period of oscillation but a series of such free periods 



