THE RADIOMETER. 
171 
square centimetre of black surface, from the radiation of a 
magnesian lamp at a distance of 150 millimetres, did not exceed 
a continuous pressure of ^ milligramme. 
On January 5, 1876, a third communication was received 
by the Royal Society, and read on February 10. The influence 
of various coatings on radiating and absorbing power is herein 
discussed, and a quantitative instrument on the principle of 
Bohnenberger’s electrometer is described, in which repulsion is 
balanced against the magnetic couple, a small magnet being 
attached to the pith, and a controlling magnet mounted ad- 
justably on the outside. 
This immediately gives the apparatus practical value as a 
photometer, by balancing a standard candle against the source 
of light to be tested. The same balance can be made between 
different colours. 
Here, for the first time, we meet with the vernacular name 
of 44 light mill ” applied to the machine, which, though probably 
imitated from the 44 mouse mill ” of another distinguished phy- 
sicist, undoubtedly had the fault of implying a theory, and cer- 
tainly affords a good instance of Bacon’s 44 Eidola Fori.” 
A fourth communication, read on the same day, details ex- 
periments on the spectrum admitted to a radiometric balance 
through a quartz window. The maximum effect is found to be 
in the ultra-red, the minimum in the ultra-violet. 
Then comes the momentous question, 44 Is the effect due to 
heat or light ? ” Tt is noted that 44 there is no real difference 
between heat and light ; all we can take account of is difference 
of wave length ; and a ray of different refrangibility cannot be 
split up into two rays, one being heat and one light.” 
Sunlight filtered through alum, glass, and water, so as to cut 
•off the whole of the dark -heat rays, produced a deflection of 
105° in the torsion apparatus ; whereas on interposing a solution 
of iodine in disulphide of carbon, the deflection sank to 2°, 
showing that the deflection previously obtained was almost en- 
tirely due to light. A number of other substances were also 
•experimented on, so as to get comparisons of action. 
Professors Tait and Dewar’s experiments, published in 
44 Nature ” for July 15, 1875, are examined by a special appa- 
ratus, and with different results from those of the original ob- 
servers, for whereas they state that rock salt is inactive to the 
light of a candle, and glass is active, the difference between 
them is found by several concordant observations to be only in 
the ratio 39 : 40. 
An extremely delicate torsion balance is lastly described, by 
means of which the radiation from a candle is measured in 
decimals of a grain ; at twelve inches off it is found to be 
0*000444 of a grain, at six inches 0*001772, or close to the 
