150 SOME KEFINJEMENTS OF MECHANICAL SCIENCE, 



a refinement most difficult in the optical surfaces upon which the 

 lines are ruled. To Mr. Brashear was given the problem of produc- 

 ing such surfaces, and notwithstanding the many difficulties encoun- 

 tered in working and refining the speculum metal plates, he has 

 made many hundred plates with surfaces either flat or curved with 

 an error not to exceed one-tenth of a wave length of light, or one 

 four-hundred-thousandth of an inch. 



As the established standards of length, which are the yard of 

 Great Britain and the meter of France, are made of metal and liable 

 to destruction or damage, Professor Michelson conceived the idea of 

 determining the lengths of these standards in wave lengths of light, 

 which would be a basis of value unalterable and indestructible. 



For the purpose of carrying out these experiments the Interfe- 

 rometer w^as constructed — an instrument which required the highest 

 order of workmanship and the greatest skill of the optician. Again 

 Mr. Brashear proved equal to the occasion, and made for the in- 

 strument a series of refracting plates, the surfaces of which were flat 

 within one-twentieth of a wave length of light, with sides parallel 

 within one second. This w^as the most difficult work ever attempted 

 in the refinement of optical surfaces. 



Professors Michelson and Morley devised a method for using the 

 Interferometer to make the wave length of some definite light an 

 actual and practical standard of length. So satisfactory was the re- 

 sult that Professor Michelson was invited to continue the experi- 

 ments at the bureau of weights and measures at Sevres, France, 

 where the standard meter, which is kept in an underground vault 

 and inspected only at long intervals, was used for that important 

 work. The final result of the experiments, which occupied nearly a 

 year, shows that there are 1,553,164.5 wave lengths of red cadmium 

 light in the French standard meter at 15° C. So great is. the ac- 

 curacy of these experiments that they can be repeated within one 

 part in two millions. Should the material standard of length be dam- 

 aged or destroyed the standard wave length of light will remain 

 unaltered as a basis from which an exact duplicate of the original 

 standard can be made. These two marvelous instruments, the Row- 

 land dividing engine and the Michelson Interferometer, show the 

 possibilities in the j^erfection of linear divisions and the standards of 

 length. 



We have recounted some steps of the progress that has been made 

 in the measurement of time, of angles, and of length, together with 

 some of the refinements in these measurements, but we are confronted 

 with the fact that notwithstanding all that has been accomplished 

 from centuries past down to the present time there are, as ever be- 

 fore, many imperfections requiring new problems in mechanical 

 science to be worked out for the further enlightenment and welfare 

 of mankind. 



