peters: measurement of small dilatations "281 



INTERFEROMETRY. — The use of the interferometer in the 

 measurement of small dilatations or differential dilatations. 

 C. G. Peters, Bureau of Standards. (Communicated by 

 S. W. Stratton.) 



The dilatometer originated by Fizeau and further developed 

 by other investigators consists of two interferometer plates 

 separated by three pins of slightly different length, made in the 

 form of a tripod or ring. When this interferometer is illuminated 

 and viewed in the proper manner, curved interference fringes 

 appear. The change in the length of the pins is determined 

 from the displacement of these fringes past a reference mark, 

 which is usually ruled in the center of the lower surface of the 

 upper mirror. The quantity that is actually determined, how- 

 ever, is the change in the distance between the two plates at the 

 reference mark, which is equal to the change in length of the 

 pins, if their behavior is identical, or nearly equal to their mean, 

 but not their individual change in length when they behave 

 differently. In fact, two of the pins may expand while one 

 contracts, or the three pins may change in such a manner as 

 to cause the fringes to rotate through 360° without causing any 

 displacement of the fringes past the reference mark. The 

 unequal variations in the length of the pins cause the direction 

 and angle of the wedge between the plates to change, which in 

 turn changes the direction and width of the fringes. This 

 was slightly noticeable in our work with three pins of steel, 

 quartz, brass, or copper, and quite pronounced with glass near 

 600° C. H. G. Dorsey^ observed this fact in his work on the 

 thermal expansion of zinc. 



The exact change in length of each individual pin can be de- 

 termined from the change in the order of interference (the num- 

 ber of light waves) between the plates at the points of contact 

 of the pins. This was accomplished without changing the 

 Fizeau apparatus in any way except to make the reference marks 

 at these points of contact. The pins were made in the form of 

 cones, about 4 mm. across the base and 10 mm. long. Three 



' Phys. Rev. 26: 5. 1908. 



