May 24, 1900] 



NATURE 



81 



rarely occurs, so a further modification is important. 

 However, by simply loading the lens fork very slightly, 

 we make the phases of the one oscillation overtake those 

 of the other as slowly as we please. During agreement 

 in phase the appearance of the bands will be that 

 already described, with the slope (let us say) downward 

 to the right. Two or three minutes after, when the two 

 phases are opposite, the slope of the bands will be down- 

 ward to the left. Between these two appearances con- 

 fusion will reign, for the rectangle is then occupied by 

 overlapping ellipses of changing eccentricity. But when- 

 ever the two oscillations are composed into a straight 

 line there is abundance of time to measure the slope of 

 the interference bands. 



We have now attained only a relative measurement of 

 intensity. But if we knew what maximum pressure 

 within the resonator produced the observed amplitude of 

 the sensitive plate, Rayleigh's expression together with 

 Helmholtz's ratio would yield us the absolute intensity of 

 the tone just outside the mouth of the resonator, which 

 we seek. This pressure we do not know ; we can, how- 

 ever, make a pretty close approximation to it. Let us be 

 content, provisionally, with an error of about four parts 

 in a thousand. Accordingly we will remove the sensitive 

 plate from the resonator, in order to substitute for it a 

 thicker plate, of natural pitch four octaves higher. Then 

 we will cork the resonator, and produce a series of pres- 

 sures within it by means of an air-pump. These 

 pressures, measfred statically with a water manometer, 

 together with the corresponding displacements, furnish a 

 table of the degree of approximation sought ; so that by 

 interpolation, when necessary, we may assign the pres- 

 sure that has caused the amplitude, P, in any particular 

 case, and thence obtain the energy of the tone in abso- 

 lute units. 



Of course, much pains must be taken to exclude all 

 disturbing vibrations from the sensitive plate, whether 



For experimentation we require a source of sound that 

 will produce a tone of great constancy and purity, but 

 one whose intensity may be varied at will between wide 

 limits. Moreover, the tone should issue from a small 

 and definitely located area. It will be convenient, also, 

 to have this instrument easily portable, so that it may be 

 moved freely even while sounding. Such a source is 



Fi<;. 3. — The source of tone, with its bax removed. 



obtained by causing a tuning fork to transmit its vibra- 

 tions to a thin iron plate, which forms a portion of the 

 walls of another spherical resonator ; for the middle of 

 one tine is rigidly connected with the centre of the plate. 

 This combination is carefully tuned to give the tone 

 required, and it is boxed so that only the mouth of the 

 resonator protrudes. The fork is driven electrically, but 



2.— The refractometer boxed and ready 

 for use. The resonator is covered 

 with felt. 



Fig. 4. — The open camera. The motor is shrouded t > prevent its sp.irking from fogginK 

 the film. Adjustment of speed is accomplished. by the aid of stroboscopic observation of 

 the disc of black and white sectors, inspected through the square of ruby glass opposite. 

 The electromagnet operates the arm which carries the shutter. 



transmitted through the air or through the floor and 

 supports. Moreover, even the waves of the tone to be 

 measured must be allowed to beat only upon the side of 

 the plate which is within the resonator. Accordingly, 

 heavy, padded boxes and piers of soft rubber are 

 employed for the refractometer, for the tuning fork which 

 carries the object-glass ; and also for the instrument 

 which produces the tone, as well as for the camera, both 

 of which remain to be described. With these precau- 

 tions, however, the result desired is very well attained, 

 as is shown by careful tests. Moreover, the constancy 

 and sensitiveness of this instrument promise to be highly 

 gratifying. 



NO. 1595, VOL. 62] 



its current is interrupted by the vibrations of a second 

 fork, the two being in relay. The intensity of the tone 

 depends, of course, upon the strength of the current 

 which drives the source-fork, and this we may vary at 

 will. Moreover, the intensity at the mouth of the source- 

 resonator may be defined in terms of the current effective 

 in producing it. These intensities are determined by 

 means of the damping factors of the arrangement. The 

 theory of this source as an independent, absolute 

 measure of intensity is an extension (Sharpe, Science, 

 1899, p. 810) of that given by Lord Rayleigh for the 

 tuning fork (Phil. Mag. 1894, vol. xxxviii. p. 365). This 

 instrument makes a very pure and effective source of 



