ON THE PITCH OF MUSICAL SOUNDS. 39 

beginning and end of the experiment, so as to give the fractional part of a revolution. The number of white lines that pass the field of view is thus easily obtained, by multiplying the number of white lines round the drum by the number of revolutions and fractional part of a revolution made by it. This, divided by the number of seconds the experiment lasted, gives the vibration number of the fork. In the hands of the inventors, this instrument has given extremely accurate results.
Another very accurate counting instrument is the Tonometer, an account of which is deferred, until the principles on which it is constructed have been explained (see Chap. xiii).
The exactness with which pitch can now be determined, is shown by the following abridged table, taken from Mr. Ellis's " History of Musical Pitch," p. 402. In the first column are the names of five particular forks, the vibrational numbers of which are given in the second, third, and fourth columns, as determined independently by McLeod with the Cycloscope, Ellis with the Tonometer, and Mayer with his modification of the Graphic method, respectively. 





Knowing the velocity of sound in air, and having ascertained the vibrational rate of any sounding body by one of the preceding methods, it is easy to deduce the length of the sound waves emitted by it. For, taking the velocity of sound as 1,100 feet per second, suppose a tuningfork, the vibration number of which is say 100, to vibrate for exactly one second. During this time it will have given rise to exactly 100 waves, the first of which at the end of the second will be 1,100 feet distant from the fork; the second one will be immediately behind the first, and the third behind the second, and so on: the last one emitted being close to the fork. Thus the combined lengths of the 100 waves will evidently be 1,100 feet, and as they are all equal in length, the length of one wave will be

