46 Messrs. G. J. Stoney and J. E. Reynolds on the Cause 



which we offer as probably a very close approximation to the 

 value of k. 



Section III. On the character of this molecular motion. 



10. But beside thus determining with very great precision 

 the periodic time of one of the motions in the molecules of the 

 vapour of chlorochromic anhydride, the study of this spectrum 

 has elicited other information about the motion, which we think 

 ourselves justified in putting upon record, although it is imperfect. 



11. It was just before leaving off work on the 28th of January 

 that we first looked at the absorption-spectrum of chlorochromic 

 anhydride, and found that it gives in the greenish yellow a spec- 

 trum of lines arranged in nearly the following pattern, 



S 



(4) 



each section of the pattern consisting of five lines, a very dark 

 one followed by a very light one, then two of medium intensity, 

 and then another very light one. 



] 2. We saw no more upon that evening ; but in thinking over 

 the pattern afterwards, it seemed to offer some hope that we 

 should be able to trace out not only the periodic time of the pa- 

 rent motion to which the lines are due, but even some informa- 

 tion regarding the character of that motion. In fact the pattern 

 seemed to suggest a very simple law of variation of intensity in 

 passing from line to line, viz. that the intensities may probably 

 be related in some simple way to the lengths of the ordinates of 

 a curve of sines raised from points which divide the interval from 

 crest to crest into five equal parts, thus 





.**T 



a. o/\ 



%. IK' 



W P) 



For example, taking the expression for the displacement-curve 

 of the original disturbance in the aether, which is* 



y=A + C 1 sin(a? + «i) + C 9 sin(2a? + a 2 )+ ..., . . (6) 



the requisite condition would be fulfilled in the simplest manner, 

 if six successive coefficients (the squares of which represent the 

 intensities of six successive lines in the spectrum) had the fol- 

 lowing values:— 



* Phil. Mag. vol. xli. (1871), p. 292. 



