LIGHT AND ITS ARTIFICIAL PRODUCTION. 291 



trurn is displaced from the red end toward the violet. If the energy of 

 certain ether waves is great enough they affect the eye as light and 

 the body emitting them is " self-luminous." We can now distinguish 

 between sources of heat and sources of light — between heat radiation 

 and light radiation. Objectively both kinds of radiation are parts of 

 the total radiation emitted by the body. In the same manner as the 

 total radiation, the partial radiation must rapidly increase in intensity 

 with increasing temperature. The light radiated by an incandescent 

 body will therefore increase much more rapidly than the temperature. 



To form a general notion of the relation between light radiation and 

 the temperature we will make use of a law recently proposed by Wien, 

 following in Boltzmann's footsteps. The law reads as follows : '^ In the 

 normal emission spectrum of a black body the energy maximum is so 

 displaced by the variations of the temperature that the product of the 

 temperature and wave length remains constant." In connection with 

 Stefan's law it states that the maximum energy in the normal spec- 

 trum is proportioned to the fifth power of the absolute temperature. 

 Basing our conclusions on both laws, we may certainly assume that the 

 light radiation increases still more rapidly with increasing temperature 

 than the total energy, since the energy maximum in highly heated bodies 

 is certainly in the visible part of the normal spectrum. Unfortunately 

 there are practically no experimental data on the relation between the 

 light radiation and the total radiation at different temperatures, and 

 we will therefore assume that light radiation as well as the maximum 

 energy in the normal spectrum is proportional to the fifth power of the 

 absolute temperature; that is, that the brightness of a luminous black 

 body — for instance, the carbon filament of an incandescent lamp — will 

 increase 2-^ or 32 times if its absolute temperature be doubled, or 3^ or 

 243 times by trebling its absolute temperature. 



Before we deduce the consequences of these laws we must discuss 

 whether they can be applied to the incandescent bodies used in prac- 

 tice. To do this it will be necessary to familiarize ourselves with a law 

 which forms the basis of spectrum analysis and which has attained 

 widespread importance. I refer to Kirchhoff's law of absorption and 

 emission of light, which states that a heated body at any temperature 

 emits only those particular rays which it absorbs at the very same tem- 

 perature. If we apply this law to bodies rendered luminous by high 

 temperatures it states that all those bodies are nonluminous, however 

 high the temperature, which either freely transmit the light rays or 

 which reflect them entirely instead of absorbing them. Highly heated 

 gases absorb no light and therefore do not emit any, as you can see in 

 the Bunsen flame. [Experiment.] Carbon absorbs a large proportion 

 of incident light even in an incandescent state. If heated to the same 

 temperature as a Bunsen flame it emits a correspondingly large amount 

 of light. All solid bodies and all metals exhibit a similar behavior. 

 These absorb to a greater or less degree waves of all wave lengths and 



