DIAMETERS OF THE STARS DAN JON". 175 



systems produces a uniform intensity. The application of these 

 considerations to the angular diameters is direct. If we have a star 

 for our source whose angular diameter we may call to, and if we 

 can cause the fringes to vary and disappear at will, we can find by 

 trial the value of 8 for which they first disappear. From the formula 

 just given we at once obtain co. 



THE! APPARATUS OF STEPHAN. 



The method suggested by Fizeau was put into practice by Stephan 

 at the Marseilles Observatory in 1873. The telescope objective of 80 

 centimeters aperture was covered with an opaque screen having two 

 apertures symmetrically placed with regard to the center (fig. 2). 



°2 



Fig. 2. — Plan of Stephan's apparatus (1873). 



A screen, shown in plan at the left of the figure, limits the portion of the objective 

 utilized to two openings, d and 2 . The fringes are observed with a powerful eye piece 

 at the image of the star E. 



Under such conditions the objective utilizes only two narrow pencils 

 of rays coming from the star. These converge at the focal plane E 

 to the image of the star. If the image is examined with a powerful 

 eye-piece the usual diffraction spot will be seen furrowed now with 

 dark equidistant bands perpendicular to the line C^CX which joins the 

 centers of the apertures. These are the interference fringes, exactly 

 analogous to those which form on the retina in Young's famous 

 experiment and of which the angular separation is 



S=X/D where X is the wave length of light and D the distance 1 2 . 



In the experiments of Stephan, the apparatus could not dis- 

 tinguish angular diameters less than 1/5 of a second of arc. Now 

 all the stars observed by Stephan in 1873 and 1874 gave very clear- 

 cut fringes. None of them could therefore equal 0".2 in diameter. 

 Probably the largest were far beneath this value. We know to-day 

 that they do not exceed 0".05. 



The interferometer method was later independently applied by 

 Michelson and Hamy to the measurement of the satellites of Jupiter. 

 As this meant working with diameters of 0".5 to 1" a telescope of 

 30 centimeters aperture was amply sufficient, and positive results 

 were obtained. 



