492 Mr Filon, On the variation with the wave-length, etc. 



But here it should be noted that, besides the limited range of 

 the observations, which are restricted to three wave-lengths and 

 do not go beyond the green 1 , Pockels' method appears open to one 

 objection, when the dispersion of double refraction is small. 



He uses a Babinet's quartz compensator: the observations are 

 consequently affected by the dispersion of double refraction in 

 quartz: it then hardly seems satisfactory to introduce, as he has 

 done, a calculated correction for this, especially when, as in some 

 of the cases he treats, the dispersion of double refraction only 

 appears in the final result through this calculated correction. 



Accordingly it seemed of interest to publish the present ex- 

 periments, which are free from this source of error, if only as a 

 confirmation by an independent method. 



I am at present undertaking a series of observations, of which 

 it is hoped eventually to publish an account, with a view to deter- 

 mining accurately, if possible, the law T of variation of artificial 

 double refraction. 



In conclusion I wish to express my thanks to Professor Ewing, 

 in whose laboratory this work has been carried out, and also to 

 Professor J. J. Thomson for kindly letting me use some of the 

 optical apparatus belonging to the Cavendish Laboratory. 



I also wish to gratefully acknowledge the kind help which was 

 given me by Mr H. C. Ramage, of St John's College, in connection 

 with the arrangement and identification of the comparison spec- 

 trum. 



1 This, I think, accounts for his negative result for glasses of densities between 

 2 and 3, the greater part of the dispersion being then in the blue end. 



