270 PROCEEDINGS OP THE AMERICAN ACADEMY 



abled to annul the effect of every sound but that which was re-enforced 

 by the resonator. Moving now the resonator along the rod, I was able 

 distinctly to mark points of maximum and of minimum intensity, which, 

 in general, coincided with the positions which these bands should 

 occupy as calculated by formula essentially similar to those applied 

 in the diffraction of light. 



In the first experiments, a number of quite serious difficulties were 

 encountered. Perhaps the most important was that due to the fatigue 

 of the sense of hearing, in consequence of which a maximum was 

 estimated before it actually occurred. By alternately opening and 

 closing the mouth of the resonator with the finger many times in 

 quick succession, however, and by taking a reading by first moving the 

 resonator in one direction, and another by moving it in the other, it 

 became possible to set it with considerable accuracy, the differences 

 being, generally, only a few centimetres. Another difficulty was met 

 with in the shape of a distinct band of interference, seeming to have no 

 connection whatever with the other bands, and following quite a different 

 law. Upon tracing it out, however, it was found to be due to an 

 interference of the direct wave with the wav^e reflected from the gas- 

 holder used to blow the pipe. Upon covering the holder with a cloth, 

 this was very much dimiuished : and, upon removing the holder, the 

 interference band disappeared altogether. There seemed to be slight 

 evidences of nodes and loops formed in tlie hall, as in an organ pipe ; 

 these, however, were very indistinct indeed, and were probably not a 

 source of error. An attempt was made to use the manometric flame 

 and revolving mirror to determine the points of most complete inter- 

 ference, but the difference in effect upon the flame was so slight as to 

 render this method entirely impracticable. 



In fact, the phenomena of diffraction can be studied only by the 

 closest attention with the ear, and the ear is certainly far more delicate 

 than any instrument of this kind that has ever been constructed. A 

 very pure note, and one of constant intensity was necessary for the 

 best results, and these were obtained by blowing the pi^je with a 

 stream of air from a gas-holder, having an arrangement for keeping 

 the pressure constant. 



Second series. The same board was set up near the middle of the 

 hall. Two hundred centimetres from its middle point, on one side, 

 was placed the pipe used in the pi'eceding experiments. On the other 

 side was arranged a system of co-ordinates within the sound shadow. 

 The method of determining the points of interference was the same as 

 in the first series ; excepting that only the bands of minimum intensity 



