272 Mr. W. H. Preece. On a New Standard of 



pure colza oil per hour, with a flame of 40 millims. ; while Draper 

 (1847), Zollner (1859), and Schwendler (1879) proposed a given 

 length or surface of platinum raised to incandescence by a current of 

 electricity as a useful standard. At the Congress held in Paris last 

 year it was suggested that a convenient standard would be the light 

 emitted by a square centimetre of incandescent platinum maintained 

 at its fusing point, but up to the present moment no one seems to 

 have devised what really may be called an absolute standard of white 

 light. The question is now being considered by a Committee of the 

 British Association. If it were possible to find a convenient, simple, 

 fixed, and uniform material which, having a resistance of 1 ohm, 

 could be raised to incandescence by 1 ampere flowing through it, we 

 should have a convenient standard, but this has not yet been 

 obtained. 



As I have said, the standard of comparison I propose is a uniformly 

 illuminated surface. I thought at. one time that we might have made 

 use of sunlight for this purpose, but I find that sunlight and moon- 

 light are even more variable than the light produced by artificial 

 means, and, after innumerable trials, I have come to the conclusion 

 that the standard which I have proposed is one very easily obtained, 

 and though not absolutely fixed, can be made as nearly uniform and 

 reproducible as any standard hitherto suggested. It is true that it is 

 of a secondary character, and dependent primarily upon a recognised 

 standard source of light ; but this primary standard remains in the 

 laboratory, while the measurement can be made in the streets, on 

 board ship, or wherever the surface of illumination may be. 



4. Having thus fixed upon a standard of illumination it became 

 desirable to devise an instrument that would enable us to compare 

 areas differently illuminated. 



Fig. 1. 



For this purpose I make use of a small Swan glow lamp giving 2J 

 candle power with the current produced by 5 volts. The construction 

 of the photometer is shown by figs. 1 and 2. The lamp (L) is fixed 



