parts of the spectrum, 

 alterations in / and /.-. 

 Lane-Fox lamp.* 



. Abney and Col. Testing. [June 19, 



The following Table I will give an idea of the 

 The observations were made a year ago on a 



Table I. 



No. 1 ray is the limit of the visible red, and is approximately para- 

 bolic, 18 watts having to be deducted to arrive at the origin of the 

 curve, and Z=311- 2. 



In No. 2 



n= 



U= 



=38-8. 

 632. 



In No. 3 



M= 



u= 



=18-3. 

 959. 



In No. 4 



7=157. 

 -= 1-48. 



It will be noticed that in the curves of rays 2, 3, and 4, the move- 

 ment of the origin from zero of energy must be very small, and is 

 therefore negligible; hence for all the hyperbolas, which differ appre- 

 ciably from parabolas, we may take the zero watts for the origin of 

 the curve. 



We further give tables showing that the results are true with the 

 visual rays as measured by ordinary photometric methods. 



Table II shows the calculated and observed intensities of three 

 visible rays compared with the same rays in a gas flame. The 

 method of observation adopted was by means of a spectrophoto- 

 meter which one of us has used during the last ten years. It con- 

 sists of two slits, SS, through which the light from the two soui 

 which have to be compared respectively pass, striking two right angle 

 prisms, P, superposed over one another as shown in fig. 1. These 



Table XIII is a still more complete record of the alterations in If and /, and 

 may be studied in connexion with this. 



