Energy in the Electromagnetic Field. 



17 



The discrepancies are within the limits of experimental 

 error, so that the theory developed above appears to be 

 substantially correct. 



As has been remarked already, the theory was also tested 

 by measurements of the positions of the minima of illumina- 

 tion in different parts of the field. Some explanation is here 

 necessary with regard to the measurements of fringes in a 

 plane in advance of the " edge " (y negative). With the 

 ordinary arrangement as described above, if the microscope 

 is moved near so that its focal plane may be in advance of 

 the " edge," we are unable to see the exact phenomena in 

 that plane, since the light has to come past the edge before 

 it can fall on the objective, and secondly, as has been fully 

 described in Mr. Basu's paper, the field is complicated by 

 the occurrence of the caustic and its accompanying fringes 

 formed by reflexion from the surface of the cylinder. This 

 difficulty was got over in the present work by turning the 

 cylinder around its axis till the desired plane of observation 

 coincided with the boundary of the polished surface. What 

 is meant may be better understood by a reference to fig. 1, 

 the process described being equivalent to cutting off the 

 cylinder along CQ and removing the lower half. It is not 

 easy, however, with this arrangement directly to determine 

 the value of y when it is negative ; and in Table II. below it 

 has been calculated from a pair of readings. The source of 

 light was a quartz mercury lamp with a green ray filter. 



Table II. 



Fringes between the cylindrical edge and the 

 source of light 



11. 



in rnuis. 



y= 



=0. 



y= -0-2 cm. 



^=-0-29 cm. 



Obsd. 



Calcd. 



Obsd. 



Caled. 



Obsd. 



Calcd. 



3 ... 





0-0169 



0-0169 



0-0082 



0-0082 



0-0062 



0-0060 



5 ... 





00307 



0-0302 



0-0165 



0-0163 



0-0123 



0-0120 



7 ... 





0-0420 



0-0417 



00240 



00240 



0-0183 



0-0182 



9 ... 





0524 



0-0522 



0-0310 



0-0317 



00235 



0<0235 



11 ... 





0-0620 



0-0619 



00378 



0-0395 



0-0290 



0-0293 



Phil. Mag. S. 6. Vol. 37. No. 217. Jan. 1919. 



C 



