436 Intelligence and Miscellaneous Articles. 



sensitive, it is not so convenient in practice as the high-pressure 

 flame of Professor Barrett. It is well known that these flames 

 are usually sensitive only to sounds of high pitch, and through a 

 limited range of pitch, this range becoming generally narrower 

 with increase of sensitiveness. During the last few years Lord 

 Eayleigh has used the sensitive flame with signal success in stud}'-- 

 ing certain analogies between sound and light. His interesting 

 lecture on " Diffraction of Sound," delivered a little over a year ago 

 before the Royal Institution *, served as my starting-point ; and I 

 am further indebted to him for special instructions without which 

 I should perhaps not have succeeded in performing satisfactorily 

 all the experiments mentioned in his lecture. As this lecture has 

 not thus far been re-published in America, a brief resume of it may 

 possibly be acceptable. 



Waves of light are so short that special precautions are needed 

 to exhibit the phenomena of diffraction. Light emanating from 

 a point and interrupted by an obstacle produces a shadow that may 

 be regarded for all practical purposes as geometric. Waves of 

 audible sound, on the contrary, are so long that -when an obstacle 

 is interposed the effect of diffraction masks that of radial propaga- 

 tion, and hence it is not usually easy to make a sound shadow 

 manifest. The difhculty in sound is not to produce diffraction, 

 but rather to limit it by using the shortest wave-lengths possible. 

 The pitch employed by Lord Eayleigh v,'as more than 20,000 

 vibrations per second, corresponding to a wave-length of less than 

 two thirds of an inch. To measure this the waves are reflected 

 from a surface arranged vertically across the direction of propaga- 

 tion, thus producing interference with the direct waves. The 

 position of the nodes and ventral segments is determined by 

 moving the reflector toward or from a sensitive flame interposed 

 between it and the source of sound. The flame flares in a ventral 

 segment and burns quietly at a node. The distance between two 

 points of quiescence is a half wave-length, from which the pitch is 

 readily computed. Knowing the wave-length, if this be small in 

 comparisoii with the diameter of an obstacle such as a disk, it is 

 possible to calculate the deflexion necessary for the meeting of 

 secondary w-aves behind it, from its opposite edges, in order to 

 produce a maximum or mimimum of intensity. In this way, as 

 much as eight or nine years ago, Lord Eayleigh repeated acousti- 

 cally the celebrated experiment suggested by Poisson to Fresnel, 

 and first performed by Arago, by which a bright point was found 

 at the middle of the shadow of a small dish. Applying the formula 

 for Huygeu's zones, an acoustic difli-actiou-grating was made by 

 which sound was converged to a focus, as if by a lens, the 

 flaring of the flame at this focus being very violent. Around 

 it, according to the theory, there should be several successive rings 

 of motion and quiescence, or, in other words, of noise and silence. 

 The first ring of noise, and the rings of silence that precede and 



* Proceedings of the Eoyal Institution of Great Britain^ Jan. 20, 1888. 



