ON THE PHONOGRAPH. 775 



pinions, so geared as to obtain a gradually diminishing rate of motion. Pinions 

 on the axles of the wheels serve for cords by which the phonograph and also the 

 recording drum may be driven at various rates of speed. This apparatus may be so 

 worked as to cause the cylinder of the phonograph to perform one revolution in one 

 hour instead of in half a second. The lever gave an amplification of seventy times. 

 and as many of the marks obtained on the recording cylinder were only about Jg th inch 

 in vertical height, it follows that the depths of many of the depressions on the wax 

 cylinder were not more than goW'h °^ an mcn - 



21. It has already been pointed out that with the view of opening out the curves 

 the recording drum was caused to move faster than the phonograph cylinder. Unless 

 this had been done, the curves of the individual vibrations would have crowded so close 

 together as to make it impossible to view them distinctly. This will be evident if we 

 look at one of the photographs (fig. 4, Plate L), representing in vertical length the one- 

 fifth of an inch. In this distance we may have from ten to twenty depressions, and if 

 these had been recorded on the same longitudinal extent of surface on the recording 

 cylinder, we would have had ten to twenty little waves in the one-fifth of an inch. But, 

 if an analysis of the curves has to be undertaken, it is not necessary to have the 

 cylinder of the phonograph and the recording cylinder travelling at the same rate. Even 

 although the latter moves faster than the former, this is of no consequence as the 

 coefficients of the harmonic series are altered all in the same ratio. It does not matter 

 what may be the form of the curve representing an individual vibration. This might 

 be either an exact facsimile of the depression on the wax or it may be drawn out by 

 causing the recording drum to move faster. Whatever be its form, if the curve is built 

 up of a series of curves that are members of a harmonic series, these constituents can 

 always be determined. 



22. Specimens of the curves thus obtained are shown in figs. 8 to 18, Plate II. It 

 will be seen that they vary much in character. 



VII. A Method of Recording Variations op Intensity of the Sounds 



of the Phonograph. 



23. Suppose a series of sound waves of gradually increasing intensity to fall on the 

 disk of the phonograph, the pressure on the disk will gradually increase and the normal 

 groove will be cut deeper. In this process each vibration will be a little deeper than 

 the one immediately before it, but the difference in depth will be very small. If the 

 increase of pressure of the note or chord lasted more than half a second, the extent of 

 surface covered by the recording point during that time would be nearly 7 inches, and there 

 might be from 500 to 1000 depressions in that distance. Suppose, now, that we recorded 

 all these little depressions by the slow moving lever method, it will be evident that the 

 gradually increasing differences in height of the little curves would scarcely be appreciable. 



