TRANSACTIONS OF THE SECTIONS, 37 
On the Combustion of Gases under Pressure. 
By Professor Epwarp Franxxianp, Ph.D., F.RS. | 
The author commenced by stating that the idea of the paper arose from obser- 
ving the reduction in luminosity when a candle or gas-flame is burnt in rarefied 
air, and the law deduced therefrom was that the diminution of illuminating power 
was exactly in proportion to the diminution of atmospheric pressure. Some years 
ago, while the author was on the summit of Mont Blane at night, he was struck 
with the want of illumination in the candles burnt in the tent in which they 
stopped for the night. He had observed similar results in other elevated regions. 
The diminution of the illuminating power was due to the reduction of atmospheric 
pressure. The commonly received opinion was that there must be incandescent 
solid substances in flames in order to produce a white light; but if they took an 
ordinary gas-flame, and placed a piece of paper with writing on it behind the flame, 
looking steadily through it, they would be able to read the writing as well, or nearly 
as well, asif the flame were not there at all, thus showing that flame was transparent. 
In following out this subject, he had been brought into contact with a number of 
flames which emitted a considerable amount of light, but which did not contain 
any solid matter whatever. One was metallic arsenic burnt in oxygen gas. It 
emitted an intense and brillant white light. Bisulphide of carbon also emitted 
avery intense light when burnt in oxygen; indeed so intense that it had been 
employed to take instantaneous photographs. This was produced without the 
, of a solid or liquid matter existing in the flame while the light was 
eing evolved. If oxygen and hydrogen were enclosed in a soap-bubble or other 
light envelope, and exploded, there was scarcely any light produced; but if they 
were enclosed in a strong vessel and exploded by means of an electric spar, an 
intense light was produced, the pressure or density at the moment of explosion 
being ten times as ereat as that in the previous case. Ignited gas emitted light in 
proportion to its density. The luminosity in ordinary flames, such as those of gas 
and candles, the author considered to be due to the presence of dense hydrocarbon 
vapours. One of the most interesting experiments shown was that of sending an 
electric spark first through air under ordinary pressure, and then through air under 
double pressure. The result was that the light of the spark due to ignition of the 
air was very much increased. The spark was sent also through many other 
gaseous and vaporized substances, showing most conclusively that the greater the 
vapour-density of the bodies the greater was their luminosity when submitted 
to ignition by the electric spark. 
On Refraction-Equivalents and Chemical Theories. 
By J. H. Guavstonz, Ph.D., FBS. 
The refraction-equivalent of a substance has already been defined before the 
British Association as the refractive index minus unity divided by the specific gra- 
vity ; and the largest generalization that has been made in reference to the matter 
is that the refraction-equivalent of a compound is the sum of the refraction-equiva- 
lents of its constituents. If this were universally true, it would be easy to deter- 
mine the equivalents of all the elements; but it is certain that some of these have 
more than one manner of affecting the rays of transmitted light. 
While investigating this subject, the author perceived that the numbers arrived 
at for different elements or compounds had a vhemical as well as physical interest, 
and a bearing on certain chemical theories. 
Thus Roscoe has given various reasons for considering vanadium as an element 
analogous to phosphorus. If so, it might be expected to have a very high refrac- 
tion-equivalent. The oxychloride of vanadium was examined and gave the num- 
ber 57°8, which, allowing 33 for the chlorine and oxygen, leaves 24°8 as the equi- 
yalent of vanadium, showing an amount of refraction totally different from that 
of the metals, and comparable only with sulphur and phosphorus. It is also, like 
these substances, extremely dispersive. 
Hydrogen, in the mineral acids, such as sulphuric and hydrochloric, has the 
equivalent 3°7 ; but in organic acids, such as acetic or citric, it gives only 1°3, its 
