136 Mr. G. J. Stoney on the Internal Motions of Gases 



only, or more on one side than on the other — such as is the 

 wing on one side only of the narrowest of the magnesium lines 

 which form the group b in the solar spectrum. 



4. Another and quite different case which probably exists is 

 where, in virtue of forces acting within the molecule, the periodic 

 time of a motion within it is subject to fluctuation, going through 

 a regular cycle of changes over and over again. In such a case, 

 instead of a line in the spectrum we should find a band of defi- 

 nite breadth, and usually with well-defined edges, as in the 

 spectra of silicon and zinc. It may be observed that this kind 

 of fluctuation, unlike that which is attributable to a disturbing 

 agent external to the molecule, seems quite as likely to affect 

 motions of long as motions of short period; and accordingly the' 

 species of bands of which it is probably the cause are met with 

 indifferently in all parts of the spectrum. Such bands are not 

 unfrequent ; and, in fact, the motions within a molecule are 

 usually so multifarious that we often find evidence in the spec- 

 trum of a single gas of the coexistence of all the various classes 

 of motions. 



5. It is a consequence of the laws that regulate the exchange 

 of heat, that the undulations of the sether excited by the motions 

 of any one particular range of periods within a body can only at 

 most appropriate to themselves a definite amount of vis viva, which 

 amount is a function of 6 and r, where 6 is the temperature of 

 the body, and r the period of vibration. Its symbolical expres- 

 sion would be of the form ff <f>(0T)dTdt 3 where dt is the element 

 of time, 6 constant, and r to be taken between the limits corre- 

 sponding to the breadth of the spectral line or band under con- 

 sideration. This puts a major limit beyond which the bright- 

 ness of any particular ray of a gas at a given temperature cannot 

 pass. But the intensities of the vibrations corresponding to the 

 bright rays may, and generally do, in various degrees fall short 

 of the maximum values thus fixed; or, in other words, the quan- 

 tity of gas experimented on is usually too small to be absolutely 

 opaque in reference to any of its rays, and is commonly very un- 

 equally opaque in reference to the several rays. This opacity, 

 and consequently the energy of the orbital motions upon which 

 it depends, is often very disproportionately affected by changes 

 in the general vis viva of the system — so much so that rays which 

 are inconspicuous as compared with others in their neighbour- 

 hood at one temperature, may become the most prominent rays 

 at a more elevated temperature, as in iron. It sometimes hap- 

 pens, too, that under circumstances which have not yet been 

 sufficiently studied, and which are but little understood, but of 

 which temperature is one, the whole character of the motions 

 changes — so that below a particular temperature they give rise to 



