126 A*NNTJAL OF SCIENTIFIC DISCOVERY. 



planetary spaces. This distinction, however, will do little toward 

 bringin^ the two results into greater accordance, as the theoretical dif- 

 ference of velocities is less than -g^Vo f tne whole, or ^ ess tnan 7( ^ miles. 



Compare with tliese conclusions of astronomy two experimental re- 

 sults on the same subject. The first attempt to obtain the velocity of 

 light, by direct experiment and test, was made by the eminent French 

 physicist, Fizeau, in 1849. The plan adopted by him for resolving 

 this abstruse problem will be readily understood from the following 

 description. If a wheel finely cut into teeth on its circumference is 

 put in rapid rotation,* a ray of light, which escapes between two con- 

 secutive teeth, will, after being reflected perpendicularly by a mirror, 

 return to strike the wheel at a different point, and either be intercept- 

 ed by a tooth or admitted at another interstice. Suppose the velocity 

 of the wheel just sufficient to bring the adjacent tooth to the position 

 whence the ray first started, in the time which the light occupies in 

 going to the mirror and returning. In this time the wheel has moved 

 over an angle found by dividing 360 by twice the number of teeth 

 which the wheel contains. Therefore the time taken by light, in go- 

 ing over a line equal to twice the distance of the mirror, is that portion 

 a second found by dividing unity by the product of the number of 

 turns the wheel makes in a second, multiplied by double the number 

 of teeth on the wheel ; the velocity of the wheel being first made the 

 smallest which will cause it to intercept the light. Such an experi- 

 ment was made in 1849, by Fizeau, the wheel being placed in a tower 

 at Suresne, near Paris, and the mirror upon a hill (Montmartre) at the 

 distance of 8G33 metres. As the wheel contained 720 teeth, and the 

 slowest velocity which produced obscuration was 12.6 turns a second, 

 it appeared that light required ^-gy^ of a second to go 8633 metres 

 and return. Hence its velocity was 313,274,304 metres, or 194,667 

 miles a second. 



Since then, M. Foucault has successfully achieved the measurement 

 of the absolute velocity of light by an experiment which admits of be- 

 ing brought within the compass of a single room. The apparatus em- 

 ployed by him embodied, substantially the principle of Fizeau's appa- 

 ratus, but was much more complicated, and so accurately devised that 

 Foucault states that the mean result of his experiments can be trusted 

 to the fraction of -g^-g-. This result gives for the velocity of light 185,- 

 177 statute miles per second ; which is less by 6336 miles than the ve- 

 locity for light usually admitted into science, namely, the velocity ob- 

 tained from the aberration of light. This discrepancy between the re- 

 sults of experiment and that of the astronomical determination, which 

 comes nearest to it, is three times greater than the variation between the 

 velocity deduced from aberration and that derived from eclipses. Now, 

 neither the velocity by Foucault's experiment nor the value of aberra- 

 tion can be charged with a possible error of three per cent., or of any 

 error approaching to this large discrepancy ; and the question arises, 

 how is the new velocity of light obtained by Foucault to be reconciled 

 with the old value of aberration ? It should be stated that aberration 

 establishes only the ratio between the velocity of light and the velocity 

 of the earth; but this ratio being established, the astronomer is enabled 

 to assign the value of the one with all the accuracy which pertains to 

 his knowledge of the other. Now if this ratio cannot be tampered with, 



