244 Prof. E. Taylor Jones on a 



the ratio of the period of the electrical oscillation to that of 



the fork is given by — -;. X is of course the length of two 



electrical waves on the plate, corresponding to a complete elec- 

 trical oscillation. Different exposures taken under similar cir- 

 cumstances also gave concordant results. Thus in the first case 

 given in the table, section 10 : three curves were measured, 

 giving the results 2354, 2354, 2352 millionths of a second. 



The tuning-fork was one of a set by Koenig, and was 

 marked as having a frequency of 768 complete vibrations 

 per second. It was not thought that the attachment of the 

 small mirror to the fork would appreciably alter its frequency, 

 and the above number was used in the calculations. 



The method just described may be used for comparing the 

 period of the fork with the interval between two successive 

 interruptions of the circuit of an electrically maintained fork. 

 In circuit with the electromagnet of the latter was connected 

 the primary of an induction-coil, the secondary of which was 

 connected to the electrometer. The curves and spots were 

 photographed as described above. Fig. 5 shows three successive 

 interruptions of the tuning-fork circuit. During this exposure 

 the secondary of the induction-coil was also connected to the 

 vibrograph (described in the next section) and was sparking 

 to the cylinder. This lowered the potential of the terminals 

 of the coil, and in order to obtain a fairly large deflexion at 

 each interruption the tension of the strip of the electrometer 

 was diminished. Much larger deflexions were produced if 

 the sparks were not passing. The frequency of the electrical 

 fork, as determined from the vibrograph record, was under 

 these circumstances 48*841, and owing to the comparatively 

 long interval between two successive interruptions, the corre- 

 sponding lengths of the two curves on the plate could be 

 determined with a very small percentage error. 



The electrical fork was actuated by a mercury break, and 

 the interruptions of the circuit did not take place with perfect 

 regularity, a long interval sometimes occurring followed by 

 a short one. In one case, for example, taking 48*841 as the 

 frequency of the interruptions, two successive waves gave 

 the values 771*2 and 765*6 for the frequency of the smaller 

 fork. The mean of six such measurements gave 767*7. In 

 order to determine more accurately by this method the 

 frequency of the smaller fork a large number of waves would 

 have to be measured, and this was not considered necessary 

 for the purposes of the present experiments. The method is 

 more suitable for examining the regularity of the interruptions 

 of the circuit of an electrical tuning-fork. 



