284 



SCIENCE 



[N. S. Vol. XLIV. No. 1130 



" fine silver sand," from which the smaller par- 

 ticles, to the extent of about one third of the 

 mass, had been sifted out for another purpose. 

 There is to my knowledge nothing critical 

 about these specifications. They are simply the 

 result of guess and circumstance, with the re- 

 sult about to be stated. 



Several resonators were tried until one was 

 found that worked properly in the position 

 shown in the figure. Along with the general 

 hiss and roar of the impact of the stream a 

 faint, fluttering musical tone could then be 

 distinctly heard when the ear was held close to 

 the mouth of the bottle. By repeating the ex- 

 periment with various amounts of water in the 

 bottle tones of various pitches could be ob- 

 tained, in every case sensibly identical with 

 the tone obtained by blowing across the mouth 

 of the bottle. 



It would seem in advance that out of a 

 helter-skelter series of impacts a group could 

 be selected having, within certain limits, any 

 given period with a sufficient degree of accu- 

 racy to set a resonator into action. Naturally 

 such a state of things could not continue in- 

 definitely. The individuals of the group 

 could be expected to get out of step, stop the 

 resonance by interference and set it going 

 again in another phase. Hence the fluttering 

 quality of the note, due apparently to the sepa- 

 rate wave-trains so set up. 



If the regular periodicity is a function of the 

 analyzer, how may two pulse-series as sup- 

 posed in the case of black -body radiation at two 

 different temperatures give rise to character- 

 istically different spectra? The answer to this 

 question seems to me now quite natural. If we 

 consider the effect, in this experiment, of vary- 

 ing the size of the constriction which limits 

 the outflow of sand, it seems probable that in- 

 creasing the outflow, by increasing the average 

 number of impacts per unit time, would cause 

 the resonator to give relatively greater response 

 (•as to amplitude or energy) at higher fre- 

 quencies and vice versa. Another condition 

 bearing on the " spectral distribution " of 

 energy in this case would seem to be the rela- 

 tive numerousness of the different-sized par- 

 ticles composing the sand; other conditions 



being equal, the smaller ones presumably tend- 

 ing on the whole to give rise to high, the 

 larger to low frequencies. This is merely 

 speculation, as the careful experimentation nec- 

 essary to show such changes has not been 

 carried out. 



The experiment as described here is scarcely 

 demonstrable to more than one person at a 

 time. It has certainly yielded large educa- 

 tional returns, to me personally at least, con- 

 sidering the insignificant outlay of time and 

 material. I am especially interested in know- 

 ing whether it is essentially new or whether it 

 has been proposed or used before. 



P. W. Cobb 



Nela Research Laboratory, 

 National Lamp Works of 

 General Electric Company, 

 Nela Park, Cleveland, Ohio 



a primary circuit key for quantitative 

 induction work 



Physiological investigation requiring either 

 the calibration of an inductorium or the use of 

 such calibrated inductorium necessitates a 



" make " and " break " key in the primary cir- 

 cuit which possesses certain qualities. Each 



