301 
The energy at any point in the wave front of the wavelets must be small 
compared to the energy at any point in the main wave, for two reasons. In 
the first place only a fraction of the energy of the original wave passes through 
the apertures. In the second place, what does get through spreads out to 
form the wavelets and thus greatly reduces the energy propagated in a partic- 
ular direction. If the speed of propagation decreases with the energy of the 
sound wave, and, therefore, with the intensity, it would seem that our photo- 
graphs should show two results: the velocity of a wavelet should be less than 
that of the main wave, and the wave front of a wavelet should not be cir- 
cular, because the energy at a point in the wavelet falls off rapidly as the dis- 
tance from the pole of the wave increases. One need not cite Stokes’s law, 
for the pictures clearly indicate a variation inintensity along the front of the 
wavelets. Yet, taking into consideration the breadth of the apertures the 
wavelets are circular, showing that the velocity of the pole of the wave is not 
greater than the velocity tangent to the grating surface. Nor does the breadth 
of the aperture, and, therefore, the energy passing through, appear to make 
any difference in the velocity. It will be noted that the photographs show 
apertures of four different sizes. 
The photographs show that the main wave and the poles of all the wave- 
lets are tangent to one another, and since the wavelets are circular, that the 
velocity of the attenuated wavelet propagated tangent to the grating surface 
is not less than the velocity of the main wave of much greater intensity. 
Physies Laboratory, Indiana University, December, 1915. 
