DEPARTMENT OF GENERAL PHYSICS. 503 



The writer's best tbauks are due to Messrs. James Powell &. Sou, of Loudon, 

 for the very generous help which they have given in connection with the pro- 

 duction of this glass and for the interest they have kindly taken in the 

 experiments. 



2. Radiation from Gas-Mantles. By J. Swinburnr. 



As the object of this paiier is to stimulate a discussion in order to elucidate 

 the subject, 1 will state one point of view and oppose some others, inviting 

 criticism, even if quite destructive; and see what happens. 



The ordinary explanation of the great luminous efficiency of the gas-rnantle is 

 that rare earths have a property of selective radiation, in virtue of which they 

 send out a larger proportion of their radiant energy in the form of light than 

 ordinary hot bodies. 



Other explanations are that there is catalytic action going on ; that cerium 

 has two oxides and oscillates between the two states, and this produces the 

 pftpcfc 



Another explanation, first given, I believe, by Ram (' Incandescent Lamp,' 

 p. l!)6), is that the Bunsen flame is really very hot, and that the mantle is of 

 such low emissivity that it gets rid of so" little power that there is little differ- 

 ence of temperature between it and the flame, and it is therefore hot enough to 

 give the light by pure temperature radiation, without any anomaly. 



One reason why the simple temperature explanation has been questioned is 

 that the temperature of the Bunsen is generally taken to be much lower than 

 it is. It is generally measured by means of platinum wires or thermocouples. 

 These can never rise to the real temperature of the flame, as they are radiating 

 and must therefore be taking in heat by conduction, in which case they must be 

 cooler than their surroundings. 



The simple temperature explanation fits the phenomena. If pure thoria has 

 low emissivity it will rise to a temperature near that of the shell of flame bathing 

 it. Having low emissivity it will then give out little light, but the light will 

 have a large proportion of visible and refrangible rays. If a very little of a body 

 with a high emissivity be added, radiation will increase, but the temperature of 

 the mantle will fall, as there must be a steeper heat-gradient to supply it. The 

 total radiation is then increased, and though the proportion which is luminous 

 will be diminished, the total light will be increased. Further addition of the 

 emissive substance increases the total radiation and reduces the temperature 

 until the light given is less even than with pure thoria. 



By selective radiation may be meant that a body at the temperature of a 

 black body emits some rays, and omits others, or that it has the power of 

 emitting more refrangible rays than a black body at the same temperature. If 

 two black bodies are in a reflecting envelope, at the same temperature, each 

 radiates to the other, and absorbs power from the other. The heat in each is in 

 a state of degradation corresponding to the temperature, and in a state of 

 equilibrium it must be radiated and absorbed by each without further degrada- 

 tion. Heat radiated from a black body into a closed space in equilibrium is 

 thus not degraded. If a body only emits the portion of the rays of high 

 frequency, though it may radiate less power or energy per second, that energy 

 must surely be of a higher grade than that of the black body at the same 

 temperature, because it can be degraded into radiation of lower frequency. If 

 that is so, this sort of selective radiation violates the second law. Emitting 

 more refrangible rays than the black body is worse still. It does not follow from 

 this that a body cannot emit rays of high frequency balanced by another batch 

 at low frequency, so that their degradation corresponds with the temperature. 

 This form of selective emissivity has not yet been invented by the advocates of 

 this theory. 



The catalytic action theory is vague. Generally catalytic action merely 

 accelerates a "change from unstable to stable equilibrium. It is difficult to see 

 how catalytic action can affect radiation, A suggestion may be made. The 



