﻿Phenomena in Photometry. 125 



different to the mean o£ those taken with the telescope in, 

 and the difference was always in favour of the redder of the 

 two lights. It need hardly be said that the difference 

 observed might be important in such work as these two 

 standards are used for. 



There is one other point that requires mention. It is, of 

 course, often necessary to reverse a photometer in order to 

 correct for any differences between the two sides of the 

 screen &c. However, for lights of similar colour, very little 

 difference is produced by doing so, as a rule. 



But when the lights differ in colour, the Lummer-Brodhun 

 behaves differently from the Jolyand Grease-Spot photometer. 

 In the case of the latter, the image the retina receives is 

 unaltered by reversing. But, with the Lummer-Brodhun, 

 the image is reversed. If, before reversing, we see a green 

 disk with a red centre, after reversing we see a red disk with 

 a green centre. 



We should, therefore, expect a much greater difference on 

 reversing the photometer in the case of the Lummer-Brodhun. 



The following figures, obtained for similarly coloured lights, 

 and for red and green lights, exhibit this : — 



Photometers 

 used. 



Lights of similar colour. 



Red to Green light. 



Ratio ( 

 1st Pos. 



)f C.P. 



2nd Pos. 



Per cent, 

 differ. 



Ratio < 

 1st Pos. 



rf C.P. 

 2nd Pos. 



Per cent, 

 differ. 



Joly 



0-93 

 0-94 

 095 



0-91 

 0-91 

 0-91 



2 

 3 



4 



224 

 1*81 

 2-10 



212 

 1-70 

 1-75 



55 

 



18-0 



Grease-Spot 



Lummer-Brodhun 



. (3) The Purkinje phenomenon has often been referred to as 

 the chief source of trouble in heterochromatic photometry, 

 but it appears to be only troublesome at very low illuminations. 

 An experiment was shown illustrating the Purkinje effect. 

 A series of coloured screens, diminishing in size from about four 

 feet square, as shown in the diagram (Hg. 3), were illuminated 

 by a 32 C.P. glow-lamp at a distance of about ten feet away. 

 AH these screens were made from the same identical red and 

 blue paper, but, even at normal illuminations, it could be 

 seen that as the surfaces became smaller the red appeared 

 brighter and brighter in comparison with the blue. The red, 



t^(o5Z 



