VG FvoL AVood on the Dispersion^ Absorption, 



In jiir at atmos])heric ])ressure the falling off of the con- 

 ductivity with time is much slower than at low pressures. 

 Thus Mr. H. A. Wilson* found the following numbers for 

 the current in amperes between two coaxial platinum cylinders 

 at the temperatures stated. The potential-difference was 800 

 Yohs and the outer tube was negative. 



Temperature 800°. 900°. 1000°. ; 1100^ 



July 6 8 xlO-'3 40 x 10-6 120 X 10-6 ' -100x10-6 



July 10 2 X „ 11 X „ 50 X „ I 140x „ ] 



July 30 -Ix „ -Tx „ 2oX „ j 8x „ i 



The greater persistence of the positive leak at high pres- 

 T?ures is probably to be attributed either to the greater 

 difficulty experienced by the induced activity in diffusing 

 away, or by the fact that a nev\^ active layer is continually 

 being reformed on the wire. T\ hich of these two views is 

 correct is a matter for future experiment to decide. In the 

 present experiments the induced activity appeared to be more 

 persistent at '3 mm. than at '0(S mm. 



The activity induced by the luminous discharge, and 

 several other points which are not definitely settled in this 

 communication, are being further investigated at present. 



The experiments were carried out in the Cavendish 

 Laboratory, and the author's best thanks are due to Professor 

 J . J. Thomson for his kindly interest and advice during the 

 course of (,he work. 



XIII. The Anomalous Dispersion, Absorption, and Surface- 

 Colour of Sitroso-dimethi/l-aniline, with a Note on the 

 Dispersion of Toluine. By It. AV. Wood, Professor of 

 JEcVperimental Plii/sics in the Johns Hopkins University ■\. 



[Plates YL-VIIL] 



THE very high dispersion of the aniline dyes and other 

 absorbing media is due, as is well known, to the fact 

 that the absorption-band lies within the visible spectrum. 

 The absorbing power of these substances, for wave-lengths 

 far removed from the centre of the absorption-band, is so 

 great that only prisms of very small angle can be used, which 

 puts a limit on the length of the spectrum which can be 

 obtained with them. 



* Phil. Trans, vol. cxcvii. p. 415. 

 t Commimicated bv the Author. 



