ON THE PHONOGRAPH. 777 



easily read tracing is obtained. In such a distance we may have one little wave represent- 

 ing the pressure of a chord lasting for one-fourth of a second, or we may have from two to 

 as many as fifteen little waves, often varying much in general character. Suppose we 

 find as many as fifteen ; then each must have lasted not more than JLth of a second. 

 Even then the ear is able to follow the individual notes, when the phonograph is listened 

 to simultaneously. This may be readily done either by listening directly to the phono- 

 graph or by connecting a telephone with the secondary of an induction coil, while the 

 current in which the variable resistance apparatus is interposed passed through the primary. 

 If, then, we hold the telephone to the ear while we look at the little pen writing on the 

 recording drum, it is easy to see that the sensations are simultaneous. Now if a note 

 of a pitch say of 300 vibrations per second lasts only j^jth of a second, it is evident that 

 only five vibrations must have occurred in that time. This shows that we can appreciate 

 a tone and decide as to its pitch if only five vibrations fall on the ear. This conclusion 

 coincides with the opinion I arrived at during last summer from a careful inspection of the 

 photographs and of the mechanically recorded curves. Of course I assume that the 

 music is being played by the phonograph in its proper tempo. If the phonograph is 

 made to travel faster, possibly it might be found that pitch might be appreciated for 

 even shorter periods. Examination of the curves shows that as a rule no " chord " lasts 

 longer than half a second. This method of recording seems well suited to the study of 

 the time relations if a series of complex sounds pour in upon the ear. 



25. If one doubts whether the movements of the recording; lever coincide with 

 the tones of the phonograph, three ways are open by which the statement may be 

 put to the test : — (1) Listen attentively with the telephone and at the same time 

 watch the recording point. The sensations of hearing and of vision for any particular 

 note are simultaneous. (2) Remove the elastic tube from the recording tambour and 

 place it in the ear and the music will be heard. (3) Lead the elastic tube from the 

 electric tambour to a recording phonograph, and a feeble record will be obtained of the 

 music showing that all the vibrations are present. In the two last experiments, as 

 might be expected, quality suffers, but the rhythm, the tempo, and the general character 

 of the tune are reproduced. Examples of such tracings are shown in fig. 5 (Plate II.). 



VIII. The Effect of Reversing the Action of the Phonograph. 



26. Long ago, with the tinfoil phonograph it was noticed that, with speech, if the 

 movement of the cylinder was reversed, there was a reversal of the sounds. I recollect 

 that in 1878 Lord Kelvin and I often repeated this experiment. Recently I have re- 

 examined this matter with the aid of a phonograph in which a reversed movement has 

 been obtained by a change in the position of the brushes of the electric motor that 

 drives the machine. It will be observed that this is the most evident way of obtaining 

 a reversed action. It will not be obtained by simply reversing the wax cylinder on the 

 india-rubber mandril of the machine, because, in the first place, the wax cylinder is of 



