PROCEEDINGS OF THE ACADEMY AND AFFILIATED 



SOCIETIES 



THE PHILOSOPHICAL SOCIETY OF WASHINGTON 



The 774th meeting was held on May 13, 1916, at the Cosmos Club, 

 President Briggs in the chair; 53 persons present. The minutes of the 

 773d meeting were read in abstract and approved. After discussion 

 the following resolution recommended by the Executive Committee was 

 unanimously carried: 



Resolved: That the Philosophical Society of Washington most heartily 

 approves of House o£ Representatives bill No. 528 (known as the 

 Johnson bill) for the adoption of the Centigrade temperature scale in 

 Government publications with the amendments as approved by the 

 National Academy of Sciences. 



Mr. C. W. Hewlett of the Johns Hopkins University presented a 

 communication on The analysis of complex sound waves and exhibited 

 and demonstrated the apparatus used in the investigation. The 

 speaker first gave an example to show what is meant by the analysis 

 of a complex sound wave. Suppose we have three tuning forks whose 

 frequencies are related to one another as are the numbers 1, 2, and 3, 

 and let us consider the sound waves passing through a point at some 

 distance from the forks. If each of the forks were excited separately 

 we should observe at the point in question three simple harmonic trains 

 of waves of definite intensities with frequencies corresponding to the 

 frequencies of the forks. If all the forks were excited together we 

 should observe a complex train of waves passing through the point, 

 and by a proper arrangement of apparatus this complex wave could 

 be analysed. If our analysis led to the conclusion that the complex 

 wave was composed of the three simple harmonic trains which we ob- 

 served when the forks were excited separately, then we should say that 

 the analysis was correct. But, as is well known, a Fourier analysis 

 could be carried out which would lead to an entirely different result, 

 and it would also be a correct analysis in the sense that the portion of 

 the complex wave analysed could be reproduced by the combination of 

 the components found by the analysis. Again suppose that the three 

 forks had frequencies whose relations to one another were the same as 

 the numbers 1, 2.1, and 3.5. The complex wave from these three 

 forks could again be analysed, and the analysis would be again re- 

 garded as correct if it gave us three simple harmonic trains of waves 

 with frequencies corresponding to those of the forks and intensities 

 which we would observe with the forks vibrating separately. But a 

 Fourier analysis in general would give us an entirely different result, 



402 



