lU ON THE APPLICATION OF INTERFERENCE METHODS 



overlapping. Notwithstanding these difficulties, the close agreement of a 

 number of olisen-ations shows that the cm-ve for the lower hne, given in 

 Fig. 1-ift, Plate III., is a close approximation to the ti-uth. Neglecting the 

 effect of a hne of feeble intensity at a distance of about 0.24 from the prin- 

 cipal line, the distribution of light in the source is represented in Fig. 14, 

 which gives for the visibility cui-ve 



V=\ a/sI? -+ F,' + 6 F, r, cos 2 irX'28, 

 in which Fj = 2 ' **'» 



and F, = 2'^' ^' cos 0.5 '280. 



Fig. 15&, Plate III., represents the results of observations on the upper 

 yellow hne, omitting some pecuharities due to the presence of one or more 

 lines of feeble intensity. The curve agrees closely with the fonnula 



V=^ a/ 3~f7+F,' + 6 F, F, cos 2 ttX/TO , 

 m which yi — ^ ' 



and Fj = 2 "*', 



which represents the ^nsibihty curve produced by two lines of intensities 

 1 : 3 and separated by 0.019 chvisions as shown in Fig. 15«. 



The gi-een mercury line is one of the most complex yet examined. The 

 constituent lines are nevertheless so fine that the interference bands are fre- 

 quently visil)le when the difference of path is over four tenths of a meter ! 



The full curve in Fig. liib, Plate III., gives the results of observations cor- 

 rected for personal equation, while the dotted ciirve represents the equation 



V = 2 ~^ ' ^'"' y/omJV + om F» + 0.28 F, F, cos 2 r.X 31 .4, 

 in which F, = 0.62 + 0.38 cos 2 zX 360, 



and V, = 0.77 + 0.23 cos 2 zA' IIU. 



This is the visibility curve con-esponding to the distribution represented 

 in Fig. 16rt. The components of the Ihie for simphcity have been assumed 

 to be symmetrical, as figiu-ed ; but the obsei-\'ations are not sufficiently ac- 

 ciu-ate to determine whether for instance each component is a double or a 

 triple hne. In this case also, as in the preceding ones, it is impossible from 



