1871.] On Double Spectra. 15 



unable to ascribe the differences observed to difference of 

 temperature. 



The conclusion drawn from this experiment appears, to 

 open up new views of the cause of double spectra, and to 

 demand further investigations which may throw light on 

 the molecular constitution of gases. We know so little 

 about the mode in which those vibrations of the atoms 

 of a gas take place by which light is produced, that it 

 is not of much use to speculate on the causes of the 

 difference of light emitted at the same temperature. We 

 know, indeed, that heat is caused by the motion of the 

 molecules of a gas, and that the faster they move the higher 

 is the temperature of the gas. If the vibrations which con- 

 stitute light are executed by the constituent atoms of the 

 molecules, it is possible that the atoms may vibrate dif- 

 ferently, although the motion of the molecules is the same — 

 that is, the light emitted may vary, although the temperature 

 is the same.* 



Further experiments may enable us to distinguish between 

 the effect of change of temperature and change of pressure. 

 It may be noted that the second carbon spectrum has been 

 obtained only from gases at pressures less than 100 m.m., 

 although the first carbon spectrum (which is produced at the 

 same temperature) can be obtained either from gases at low 

 pressure or at high pressure. It is understood that Messrs. 

 Frankland and Lockyer are engaged in experiments which 

 will throw light on this subject. 



* If this supposition is corredt, increase of temperature ought to cause the 

 lines of a spedtrum to expand, since, in the motion of a molecule. towards the 

 observer, the refrangibility of the light it emits ought to be increased, and in 

 the opposite motion ought to be as much diminished, The expansion which the 

 lines of hydrogen undergo under increase of pressure is, however, too great to 

 be accounted for by any possible consequent increase of temperature. 



