TRANSACTIONS OF SECTION A. 425 
Jodine . : : : : . 1 centigramme. 
Potassium iodid : : : . 2 centigrammes. 
Distilled water ; : ; eelOkerc: 
The cell, filled with this solution, to be held between the eye and the photo- 
meter whilst the observation is made, in order to render each light of approximately 
the same colour. When using the lamp as a standard of quantity the loop of the 
filament should be vertical and its plane at right angles to the photometer screen. 
It will be seen that by this plan a lamp of any ‘ quantity ’ may be standardised, so 
as always to radiate the same ‘quality ’ of light. 
10. On the Dependence of Total Radiation on Temperature. 
By Sir Wittiram Siemens, D.C.L., FBS. 
On April 25, 1883, the author presented a paper bearing the same title to the 
Royal Society, in which he developed a method of determining the total radiation 
and the temperature of a metallic conductor traversed by an electric current by 
measuring that current and also the electric potential between the terminals. 
In an article appearing in the ‘ Philosophical Magazine’ of September 1883, by 
Captain Abney and Colonel Festing, these authors admit the method to be one of 
‘great promise,’ but consider it defective for two reasons, viz., that 
(1) Platinum (the conductor employed in some of the experiments) is not black 
at ordinary temperatures, and 
(2) Much of the energy must have been dissipated by convection currents. 
They proceed to describe a modification of the same method, substituting car- 
bon filaments in vacuum—such as Swan or Lane Fox lamps—for the metallic con- 
ductors, with a view of avoiding the difficulties just stated. Following up the 
same modified method, Captain Abney now proposes to base upon it a method of 
fixing a standard of white light. 
Sir William Siemens takes exceptiun to this proposal, maintaining in the first 
place that the objections urged against his method admit of explanation, and in the 
next that the proposed substitution of carbon for metallic conductors would 
introduce questions of great difficulty. 
The objection raised against platinum wire that it is not black when cold 
would be easily met by the substitution of platinised platinum wire, and, in testing 
the black platinised against the bright platinum wire, an interesting comparison 
between the radiation from a black and bright surface through a long range of 
temperature could be established. In exposing the wire under examination to the 
atmosphere convection currents were no doubt set up which went in deduction of 
total radiation. To avoid these he had suspended his wires within exhausted re- 
ceivers, but found that the atmospheric density made no appreciable difference in 
the result. When the gaseous pressure was reduced below that of a millimetre of 
mercury, the loss of heat otherwise than by radiation was observed to increase on 
the contrary, pointing to the fact, previously determined by Mr. Crookes, that 
rarefied air is a conductor of heat. 
But, supposing that the rarefaction within the bulb of an incandescence lamp 
exceeds the limit at which conduction takes place, losses by convection currents 
must nevertheless take place, exceeding those from the unprotected wire, because 
the glass bulb itself absorbs a large proportion of both the low heat and the ultra- 
violet rays—as evidenced by its elevated temperature—which heat was communi- 
cated to the air by convection currents. It must be borne in mind that the surface 
of the bulb exceeded that of the ignited carbon thread nearly a hundredfold, giving 
rise to increased loss by convection. 
The substitution of carbon for metallic wire was, moreover, objected to on the 
-ground that although the electrical resistance furnished a safe indication of tem- 
perature in the case of metallic conductors, carbon was known to be affected 
very irregularly in the opposite sense. Those physicists who had endeavoured 
to establish a law of dependence between temperature and conductivity of carbon- 
