PHYSICS. 461 



Weber has described a photometer depending upon equal visual acutc- 

 ness, the two sources of light which are to be compared being made to 

 illuminate two similar halves of a glass plate on which are photographed 

 concentric circles very close together. The advantage claimed for this 

 method is the facility which it gives of comparing a difl'iise light with 

 a standard, or even two lights of different colors with one another. In 

 the latter case the observation is facilitated by the use of a red glass. 

 {Wicd. Ann., xx, 32C; J. Phys., March, 1884, II, ill, 143.) 



Wikl has shown that his polarizing photometer may be converted 

 into a spectrophotometer by interposing an Amici i^rism between the 

 double-image prism and the polariscope. The slit is placed close to the 

 terminal faces of the total-reflection prisms. A collimating lens is 

 placed against the Foucault prism. Under these conditions each of 

 the sources of light shows a channeled spectrum, the bands being alter- 

 nate in the two. Hence when the spectra are superposed, the fringes 

 disappear, for one particular portion of the spectrum. A slit attached 

 to the eye-i)iece, movable by a micrometer screw, serves to limit the re- 

 gion in which the measures are taken. {Wied. Ann., xx, 452; J. Phys., . 

 March, 1884, II, iii, 142.) 



Crova has pointed out that the complete determination of the photo- 

 metric value of a powerful light requires : first, a comi)arisou of two 

 lights of differing colors ; second, an estimation of the color by means of a 

 numerical factor; and, third, the determination of the photometi'ic ratio 

 between a very intense light and a relatively' feeble standard. The first 

 l^oint he solves by the use of a solution consisting of 22.321 grams of 

 ferric chloride and 27.191 grams crystallized nicliel chloride, dissolved 

 in distilled water to a volume of 100 c. c. at 15°, which is i)laced between 

 the eye and the screen. A thickness of 7""" of this solution allows rays 

 of wave-length 630/f to 534// only to pass, the limiting value being 580// 

 which is the best for solar photometry. The second point is attained 

 by making two successive determinations by means of the spectropho- 

 tometer — one through the above solution, giving the intensity r<itio; 

 the other through a red glass colored with cuprous oxide, which allows 

 rays from 726/< to 752/< to pass, and which gives a ratio as much be- 

 low the former one as the light is whiter than the carcel. The quotient 

 of the first ratio by the second is a constant which fixes the color, 

 having a higher value as the light is whiter, and is equal to unity when 

 the light has the same color as the carcel. For an incandescent lamp 

 it varies from 1.05 to 1.23, and for the arc 1.5 to 1.7. The third i)oint 

 requires a special photometer. {G. R., December, 1884, xcix, 1 ()(!".) 



Crova has subsequently described this instrument, which he calls a 

 difiusion photometer. On a plate of ground or of opal glass, or on a 

 Foucault screen, acting as a difluser, a uniform luminous field is thrown, 

 the incident rays falling normally. Each of the i)oiiits of the jdate may 

 be considered as a lumiuoiis source, and sends back of the screen an 

 amount of light depending on the character of the screen and varying 



