398 Messrs. K. W. Forsyth and R. J. Sowter. 



venture to think much better suited for obtaining good photographs, 

 though the drum enabled us to determine the frequency of the 

 oscillation more easily. When the drum was used, it was incon- 

 venient to take more than one photograph on the same film, and 

 moreover, it was impossible to use plates, which in some respects have 

 advantages over films. 



Having satisfied ourselves by means of the drum, that the frequency 

 of the curves we obtained corresponded to 64, we had recourse to 

 a sliding plate arrangement fitted with an automatic shutter. The 

 sliding piece is capable of carrying 3 feet of plate 2 inches in width, 

 and the shutter is so arranged that any 6 inches of the plate can be 

 exposed at will. With one filling of the slide, it is therefore possible 

 to take six photographs. 



Our first effort was to obtain photographs of the steady bands when 

 no sounds were produced. We have taken many of these, and 

 reproduce one of them in (Plate 6) fig. 1. 



Fig. 2 shows a photograph taken when a 64 fork was sounding 

 rather loudly. 



Fig. 3 shows an extraneous disturbance produced by the 

 slamming of a door. It is evident that the vibrations are compound 

 and, in part at all events, forced. We then proceeded to obtain 

 the difference-tone. The frequencies of the two notes used were 256 

 and 320, and these were produced as by Riicker and Edser with a 

 Helmholtz double wind siren. The 12 row of holes was tuned to a 

 256-fork, and then to produce the 320-note the 15 row of holes was 

 opened. 



Fig. 4 shows a photograph taken when the 256-fork was sound- 

 ing, and when the siren was sounding in unison with it, the 12 

 row of holes being open. It will be seen that the bands are steady. 

 There is a slight vibration present, probably due to the disturbance 

 produced by the blowing of the bellows. 



Fig. 5 shows a photograph taken immediately after the foregoing 

 one, upon the same plate, and under exactly the same conditions with 

 the single exception, that the 15 row of holes was opened and the 

 320-note therefore sounding in conjunction with the 256. The 

 objective reality of the difference-tone is clearly proved. This 

 experiment has been repeated on several occasions and with different 

 notes, and we have obtained many photographs demonstrating the 

 vibratory motion which is given to the bands by the difference-tone. 

 In all cases, we have proved that the separate notes sounding alone 

 produced no effect upon the bands. 



Fig. 6 (Plate 7) is another photograph of the effects produced by 

 a difference-tone. 



It will be observed in these two photographs, that although the 

 chief effect is that due to the sounding of a 64-note, there is super- 



