362 PROCEEDINGS OP THE AMERICAN ACADEMY 



Suppose a body capable of vibrating a times per second for its funda- 

 mental be struck h times per second ; then will its rate of vibrating be 

 interfered with - times. If b be less than a, then will there be a cer- 







tain number of these vibrations made per second. If b be equal to a, 

 then after the first impact the body will vibrate in its own period with 

 increasing amplitude and without interference. If b be greater than 

 a, then will interferences take place in all phases of the vibrations, and 

 the body will not make any characteristic vibrations. That is, its 

 fundamental rate will be destroyed. If the body can vibrate in any 

 harmonic series, some of these harmonics might be present associated 

 with such irregular forced vibrations above its fundamental number, 

 and a spectrum of such a body would consist of such shorter waves. 

 It would apparently be moved towards the blue end. If then the 

 light-giving molecules of the sun have either so short a free path, or the 

 velocity between impacts is so great, as to insure that the number of 

 impacts per second is comparable with the vibrating rate of the mole- 

 cules, one ought to expect that the fundamentals would largely be 

 destroyed, and therefore could have no representations in the spectrum, 

 while a colder body like the moon, with a vastly less molecular velocity, 

 might have an appreciably longer spectrum. 



