324 INFRA-RED EMISSION SPECTRA. 



the molecular speeds. The "electrical temperature" is proportional to 

 the mean square of the ionic speeds. In a strong electrical field, e. g., 

 spark between electrodes, the ionic velocities will be distributed in a 

 very different manner from that prescribed by the Maxwell-Boltzmann 

 Law, and the "electrical temperature" may differ very widely from the 

 "thermal." The arc is a region of high thermal and low electrical tem- 

 perature ; the spark has a low thermal and a high electrical temperature. 

 Of course, as the criticism can and has been made, the term "electri- 

 cal temperature" has no meaning, if taken in the sense that the word is 

 ordinarily used. I am not aware that the author meant it to be thus 

 taken. If by "electrical temperature" is meant the temperature to 

 which a body would have to be raised (by thermal means) in order to 

 cause it to give out such light as is emitted in the spark and in the 

 vacuum-tube during the electrical discharge, then the term is not mean- 

 ingless. If by "electrical temperature" is meant the temperature of the 

 ion (Stark equates the kinetic energy of the ion to its temperature), 

 then one would hardly expect to measure at such a temperature by 

 ordinary methods. Our experiences in using a platinum thermoele- 

 ment in finding the temperature of a Bunsen flame suffice to warn us 

 against such a fallacy; for, if the wires be fine enough, the platinum 

 will melt, showing that certain parts at least of the flame are at a much 

 higher temperature than indicated by the heavier wire which will not 

 melt. 



To return to the theory, there is considerable evidence for believing 

 that the chemical atom or molecule is composed of positively and nega- 

 tively charged particles called ions. These oppositely charged parti- 

 cles are equal in amount, since the molecule as a whole is electrically 

 neutral. It is possible to separate from the molecule a minute particle 

 which carries a negative charge, and thus leave an equal positive charge 

 on the remainder of the molecule. The negatively charged particle, 

 called "corpuscle," "electron," or "electronion" carries a charge equal 

 to that of the hydrogen ion in electrolysis, as shown by Townsend," 

 while its mass is about o.ooi as great as the hydrogen atom." This mass 

 and charge of the electron is independent of the kind of matter from 

 which it comes, and is invariable at low pressures. The mass of the 

 positive ion is much larger, since it is always associated with the atoms 

 of the gas in which it occurs. It is the residue after one or several nega- 

 tive ions (electrons) have been separated from a neutral atom by col- 

 lision with a rapidly moving electron. This state of affairs is brought 

 about whenever an electric field is produced, as, for example, the differ- 



^ Townsend : Phil, Trans., 193, 153, 1900. 



2 Thomson: Phil. Mag. (6), 5, 346, 1903; Wilson: Phil. Mag. (6), 5, 429, 1903- 



