﻿296 On the Cause of the Interrupted Spectra of Gases. 



gregate the appearance of patterns which often resemble the 

 flutings on a pillar. When these spectra are more carefully ex- 

 amined, it is probable that the whole series of lines occasioning 

 one of the fluted patterns will be found to be the successive har- 

 monics of a single motion in the molecules of the gas. It may 

 readily be shown that such patterns as are met with in nature 

 may in this way arise. For this purpose it is only necessary to 

 make some suitable hypothesis as to the original undulation im- 

 pressed by the gas upon the sether. Thus, if the law of this 

 undulation were the same as that of the motion of a point near 

 the end of a violin-string, and of a periodic time sufficiently 

 long (as, for example, two million-millionths of a second), this 

 undulation, when analyzed by the prism, would give a spectrum 

 covered with lines ruled at intervals about the same as that be- 

 tween the two D lines, and of intensities varying so as to become 

 gradually brighter and then gradually fainter several times in 

 succession in passing from line to line along the spectrum. 

 These alternations would give a fluted appearance to the spec- 

 trum ; and from appropriate hypotheses as to the original vibra- 

 tion, all the patterns met with in nature would result. Possibly 

 it may prove to be practicable to trace back from the appearances 

 presented within the limits of the visible spectrum to the cha- 

 racter of the original motion to which they are all to be referred. 

 But, however this may be, it will be easy in a spectrum of this 

 kind, in which we have a long series of consecutive harmonics, 

 to determine at least the period of this motion ; and it is in the 

 examination of these spectra that the most easily obtained results 

 may be expected. But the necessary observations are at present 

 almost altogether wanting. The only case in which the author 

 had been able to arrive at any result was that of the nitrogen 

 spectrum of the First Order, observed by Pliicker. It would 

 appear from his observations* that the more refrangible of the 

 two fluted patterns observed by him is due to a motion in the 

 gas having a wave-length of about 0*89376 of a millimetre, 

 which corresponds to a periodic time of three Xllth-seconds, 

 one of the .flutings consisting of the thirty-five harmonics from 

 about the 1960th to the 1995th. 



This result, however, does not command the confidence which 

 the preceding determination of one of the periodic times in hy- 

 drogen does; but it will suffice to show the character of the 

 much easier investigation which has to be made in the case of 

 gases which produce spectra of the First Order. 



* Philsephical Transactions for 1865, p. 7, § 16. 



