of Light and its Theory. 252 



(O «) — 0,00002422 

 (D a.) = 0,00002175 

 (E «) = 0,00001945 

 (F w) = 0,00001794 

 (G w) = 0,00001587 

 (H u) =z 0,00001464 



From the equation (III.) it will appear that the value of r in 

 some degree depends on y, that is to say, on the distance at 

 which the image of this phenomenon is formed, and that the 

 light, therefore, thus modified after the principles of interfer- 

 ence does not proceed in a perfect mathematically straight 

 line but in a curved line. The equation for this curved line 

 is developed without approximation, in this form, 



(IV.) a'2 [4 £2—4 (g . sfn <!+ i;w)=4 y* (s . sin tf +_vw) ^ ' 



-j-[]g2 — (s .sine +1/^)2] g , sin (Tjf vw)* 



Since these phenomena can only be observed with a telescope, 

 which, if the divergences are to be determined with some ac- 

 curacy, must have great power, and cannot therefore be very 

 short, then y is in comparison with w and g very large. In 

 my instrument the distance of the micrometer thread to the 

 systems of lines is — 21,43 inches. If y, in comparison 

 with w, were not very large, the relative value of r and y 

 might be found somewhat different ; but if in the equation 

 (III.) y is once set down J, and then two or three times as 

 large as it really was in my experiments, then the same value 



obtained in the eighth decimal place somewhat larger than when 'they are 

 obtained from the experiments with finer systems of lines upon glass. This 

 small difference can only be looked for in the less accurate determination 

 of the value of «, which I am inclined to think more accurate with the 

 finer systems of lines. It is unnecessary to remind the reader that the 

 precision in regard to the determination of the value of e has its limits to 

 whatever extraordinary a height it may reach, with the aid of those means 

 which I use ; and I have hopes of extending them still further. 



* The equations (III.) and (IV.) are developed for the case when the 

 incident rays may be considered as parallel with each other. Where the! 

 distance of the luminous point in comparison with » is not very large, 



then in both equations « sin. <r is to be placed instead of ^iiHlA____?_i 



^ cos. ^ /3 



where /3 is expressed by the equation yS = '—. In the latter «" de- 

 notes the 4istance of the luminous point from the system of lines 



