( 415 ) 

 which vahie may also be derived directly, if we equate the magnetic 

 displacement . cu R after Voigt with that resulting from the elemen- 

 tary theoiy. The dispersion of the magnetic rotation expressed hy 

 this formula is tiie same as that resulting from Becquerel's ^) relation 

 and found by Iiim to be confirmed in the case of carl)oii disulphide 

 and creosote. 



The relation found for q enables us to compute — as soon as \\q 



in 



\u\o\y the rotation constant q and the dispersion — of a substance 



dX 



for the same wavelength P.. For we \\{wq 



e 



_'1V d). 

 m A dn 



We shall make the calculation for some substances at a value of 

 ;. =z 589 mi. The rotation constants r being usually expressed in 

 minutes Ave have 



2;t 



o 



360 X 60 

 and hence we tlnd 



e 2 X 3 X 10^" 2^ dX dX 



7n- 589 36Ö^^^X^'^=^^-^^X^^^X^^- 



1. Air (100 KG., 13°.0). I have found ■') r = 553.10-". Perreau ') 

 finds for the refractive index at (1 atm., 0° C.) 



X— 644, n-nn = 85.10-« 

 538 88.10-« 



dX 

 whence — = 0.(35 X iO« and 0.58 X i<)% t>n an average O.Hl X J0\ 



Supposing 'II — 1 proportional to the density, it follows that for air 

 (100 kilogram, 13°.0 C.) r///r/>/ = 0.(M8 X iO" and we find: 



- = 2.96 X 553 X 0.648 X 10^ = 1.06 X 10'. 

 m 



In the same Avay is found for: 



2. Carbon dioxide (1 atm. 6°.5). r = 8.62 X 10-' 



dX 



— (1 atm., 0°) =3.42 X iO" 



dn ^^ 



(1 atm., 6°.5) =3.50 X iO' 



') Becquerel. G. R. 125 p. 679. 



2) SiERTSEMA. GomiTi. Lab. Leilen. Suppl. N'. 1, p. 86; Arch. Néerl. (2) 2 p. 376. 



3) Perreau. Ann. de Ch. et de Ph. (7) 7 p. 289, 



