330 PROCEEDINGS OF THE OHIO ACADEMY OF SCIENCE. 



at first by Professor Henrici, put into a commercial form by a 

 Swiss instrument maker, and somewhat rearranged by Dr. 

 Miller to increase the number of partials from ten to thirty, fur- 

 nishes a complete and precise analysis of a curve in a few min- 

 utes, thus making it possible without undue labor to analyse a 

 great number of photographs. 



In the second place. Miller has investigated carefully the 

 errors introduced into the curves by the recording apparatus 

 itself. The oscillating mirror, the cavity in which it is placed, 

 the responding diaphragm, the magnifying horn, have each their 

 own natural periods of vibration, and will respond more readily 

 and loudly to sounds or partials which are nearer to these pe- 

 riods than they can to other sounds. The errors thus introduced 

 into the curve are so great that it is hardly stating it too 

 strongly to say that an original sound-curve, either photo- 

 graphed or impressed upon a phonograph roll, is of little or no 

 value for study until it has been freed from these adventitious 

 effects. Miller spent some years of work in determining the 

 errors due to his instrument, and in devising methods to minimize 

 or correct them. The result has been a substantial advance in 

 our knowledge of the nature and composition of the sounds of 

 various musical instruments, including the human voice, and the 

 characteristics of the different vowels. 



Without doubt, however, the most important physical de- 

 velopment with which Ohio has been connected, and one which 

 is deeply concerned with our general theme of radiation, is con- 

 tained in the remarkable group of researches which is based on 

 the Michelson interferometer. Seldom or never before has such 

 important and varied work been performed with a single instru- 

 ment. It was invented by Michelson in 1881, to investigate the 

 hypotheses upon which are based Fresnel's explanation of the 

 aberration of light. These hypotheses are two : 



1. That the luminiferous ether, with reference to the earth, 

 is at rest, the earth moving through it without disturbance. 



2. That in dense optical media, however, the ether is car- 

 ried with the medium, not completely, but to an amount depend- 

 ing on the refractive index of the medium. Michelson's well- 



