CHAPTER IV. 



THEORETICAL. 



It remains for us to consider the theoretical side of this subject. 

 Experimental observations always have some value. This is not always 

 true of theories which are built, more or less, upon hypotheses and must 

 stand or fall with them. For example, nowadays one no longer con- 

 siders spectra to be due to molecular or atomic vibrations ; the divisibil- 

 ity of the atom into smaller electrically charged particles, called "ions" 

 or "corpuscles" must be assumed to account for observed facts. 



The foremost and most conservative in propounding such a theory 

 is J. J. Thompson.^ J. Stark^ is more daring in that he classifies these 

 "ions," and gives the functions that each class has to perform. 



Before considering these theories it will be well to distinguish be- 

 tween several forms of radiation, since every body gives out some form 

 of radiation, in the form at least of heat waves of great wave-length. If 

 a body can give out radiation continuously without changing its nature 

 it is called a pure thermal radiation.^ If it can not continue to give out 

 this radiation indefinitely, without changing its nature, even when the 

 temperature is kept constant {e. g., fluorites which emit light on heat- 

 ing), then the radiation is termed luminescence. The vacuum-tube is 

 thought to give out light by "electro-luminescence." 



The vacuum-tube is always quite cool in comparison with the arc. 

 this brings us to the question of "temperature." What do we mean by 

 the temperature of a body, particularly with the two forms of radiation 

 just mentioned? Both emit light; nevertheless, if we were to heat a 

 piece of iron until it felt as hot to the touch, or until the mercury in a 

 thermometer expanded to the same height, as it does for the vacuum- 

 tube, we know from ordinary experience that the iron would not emit 

 light. "Temperature" is something we are supposed to measure by 

 ordinary means. The word itself is used carelessly and is an endless 

 source of discussion. From a thermodynamic standpoint the mean 

 value of the kinetic energy of a molecule is the mechanical measure of 

 heat and temperature." To anticipate a little what is to follow. Stark uses 

 the terms "thermal" and "electrical" temperature. The purely "thermal 

 temperature" for any one gas is proportional to the mean square of 



1 Thomson: Conduction of Electricity Through Gases. 



^ Stark; Elektricitat in Gasen. 



^See Drude's Lehrbuch der Optik, p. 452. 



* See Meyer's Kinetic Theory of Gases. 



323 



