in the Hilfner Type of Spectroplwtometei 



48: 



sent back for repair. The opportunity was taken of replacing 

 the rhomb, after which it gave the correct reading, although 

 I have no actual record of the final test reading. 



Spectrophoto- 

 . meter number. 



Density of 

 standard glass 

 at W.L. 5550 

 as given by the 

 spectrophoto- 

 meter. 



Error due to 

 polarization by 



the Hiifner 

 rhomb and dis- 

 persion prisms. 



(calculated). 



Corrected 

 reading. 



1 



1-81 



1-92 



1-92 



[ 1-84 

 J 



1 V88 

 1-83 



1-86 



1-86 



+ '05 

 -•05 

 -•05 



+ ■05 







+ •05 





 



1-86 



1-87 

 1-87 

 1-89 



1-88 

 1-88 



1-86 



1-86 



2 



3 



3 

 (nieol rotated 

 through 90 °) 



3 

 (new rhomb) 

 4 





6 





One of the great sources of error in polarization photo- 

 meters is the difficulty of obtaining pure plane-polarized light, 

 and to keep it plane-polarized so that when it arrives at the 

 second nicol it can be entirely extinguished. The extinction 

 of the last two instruments was so good, that by rotation of 

 the second nicol from the zero position to the 90° position, 

 the intensity varied from about 1 to 1/10,000. Imperfect 

 annealing of the optical train and strain by clamping or 

 cementing, or flaws in the first nicol are the chief difficulties 

 met with — they give considerable but not insurmountable 

 difficulties, and are readily detected in the instrument. They 

 all tend to make densities appear too high. 



(B) The use of the instrument as a polarimeter was first sug- 

 gested by me, I believe, to Dr. K. C. Browning in conversation. 

 Some while after that he commissioned Messrs. Hilger to 

 undertake the construction of one of these instruments with 

 the suggested polarimeter addition, which consists merely in 

 the extension of the telescope support sufficiently far to enable 

 one to interpose between the dispersion prism and the second 

 nicol the medium of which one desires to measure the optical 

 rotation. 



The readings are taken by putting into the top beam a 



