248 



Velocity of White and of Coloured Light. [May 19 



The correction to vacuum is + 54 miles a second. The correction 

 for rate of clock to a mean solar time is + 52 miles a second. 



The final results for the velocity of the light from an electric lamp 

 in vacuo is 187,273 miles a second, or 301,382 kilometres a second. 



Using Struve's constant of aberration 20"'445, we obtain for the 

 solar parallax the value 8"' 77, and for the mean distance of the sun 

 93,223,000 miles. 



On February 11th, 1881, the reflected stars were seen to be coloured, 

 one reddish, the other bluish. The particular colour of a particular 

 star depended upon the speed of rotation of the toothed wheel. That 

 star which was increasing with increase of speed of the toothed wheel 

 was reddish, that one which was diminishing with increase of speed 

 was bluish. This seems to be caused by the fact that blue rays travel 

 quicker than red rays. 



A number of tests were made to judge of the accuracy of this con- 

 clusion, and they confirmed it. In the final arrangements, the 

 electric light was acted upon by a bisulphide of carbon prism, and part 

 of a pure spectrum was used. Differential measurements were then 

 made to find the difference in velocity of rotation of the toothed wheel r 

 required to produce equality of red and of blue lights. The most con- 

 venient method was to use a driving weight slightly in excess of that 

 required to produce equality of the light, then to fix to the pulley 

 carrying the weights one end of a piece of stout india-rubber tubing, 

 the other end being fixed to a point above. This gradually diminished 

 the effective driving weight. The equality of red lights was first 

 noted, the colour of the light was changed, and the interval of time 

 until the blue lights were equal was measured. The rate at which the 

 india-rubber diminished the speed was afterwards measured by the aid 

 of the chronograph, and thus the difference of speed determined. The 

 mean of 37 determinations in this and other ways gave the result that 

 the difference in velocity between red and blue lights is about 1*8 per 

 cent, of the whole velocity, blue travelling most rapidly. 



The general conclusion seems to be supported by a comparison of 

 the velocity of light measured by M. Cornu and Mr. Michelson, where 

 the source of light usually employed is taken into consideration. 

 These are the only accurate measurements of the velocity of light 

 hitherto published. They give us the following results : — 



MichGlson's research .... 

 Cornu's „ .... 

 The present „ .... 



Usual source of light. 



Telocity in kilo- 

 metres a second. 



The sun near horizon. . 



299.940 

 300,400 

 301,382 



