332 SECTIONAL TRANSACTIONS.—A. 
Monday, September 28. 
Prof. P. Zreman.—Ionised Noble-gas Spectra and the g-sum Rule. 
The spectra of ionised neon, argon, krypton and xenon under magnetic influence 
have been investigated by Mr. Bakker, working under my direction in my laboratory. 
A test was made of the so-called g-sum rule due to Pauli. 
Prof. R. W. Woov.—WNotes on Recent Investigations in Optics. 
1. Absorption Spectra of Solutions of Coloured Salis and Dyes in Liyuid Ammonia.— 
Certain coloured metallic salts are freely soluble in anhydrous liquid ammonia. The 
absorption bands are much narrower and blacker than in water solutions. The same 
is true for many aniline dyes. The broad band of continuous absorption in the ultra- 
violet of potassium permanganate breaks up into twelve components in ammonia 
solutions. In some cases the solid residue obtained by evaporation is insoluble in 
ammonia. 
2. Selective Temperature Emission.—Thin rods of fused quartz are nearly non- 
luminous in the Bunsen flame, owing to the low value of the absorption coefficient. 
Heated to a higher temperature in the oxygen-coal gas flame, the emission of white 
light sets in suddenly, there being no ‘red-hot’ stage. This is due in part to the 
drift of the ultra-violet absorption band towards the violet, and in part to the shift 
of the region of the maximum of the emission curve towards the violet with rising 
temperature. The addition of a trace of neodymium to the quartz causes the emission 
of a remarkable spectrum of bright bands with almost black intervals between them. 
3. Recent photographs of hyperfine structure of nine mercury lines with the 
40-foot grating spectrograph, with some comments on the probable origin of a type 
of ‘ ghost’ not previously discussed, which simulates fine structure. 
4, Exhibition of stereo photographic models of electron motion in Stark effect. 
Prof. Dr. Martin Knupsen.—Radiometer Force. 
A long, narrow and thin band of platinum, black on one side, bright on the other, 
was placed in a large glass vessel containing gas at a low pressure p. The band could 
be heated electrically to a higher temperature T, (absolute) than the temperature 
T, of the vessel. If the pressure of the gas on the black side of the band be denoted 
by p‘ and on the bright side by p’, then p’-p” is the radiometer pressure. The only 
reason why p’ is greater than p” must be that the coefficients of accommodation a are 
different on the two sides of the band. 
The measurements at various pressures p have shown that the radiometer force 
under all circumstances is dependent on a, whence it follows that this quantity must 
always be considered in theories bearing upon radiometer forces if a is not properly 
eliminated. 
From the kinetic theory of gases I have calculated the following expression, which 
should hold good for small values of p and T,-T) 
T, ST Ws 
4T, 
where a; and a” are the values of a on the black and on the bright sides respectively 
of the band. The index ¢ denotes that the translational energy only of the molecules 
is concerned. From this equation a;’-a;” is found, since all other quantities have 
been measured. 
The apparatus was also used for measurement of the heat transfer from a black 
platinum band and a bright one to the surrounding vessel, and from these measure- 
ments the values of a, viz. a’ and a”, were determined for the two surfaces. In the 
values thus found both the translational and the internal energies of the molecules are 
concerned. 
The experiments gave the following values : 
(as’ — a;”) 
p= HB 
for hydrogen by heat conduction . - a—a"=0,420 
B s »,Tadiometer force . . ata” =0,415 
for helium by heat conduction : . a -a" =0,498 
iy +> 3, radiometer force F . ag’—at’/ =0,512 
