Arc and Spark Spectra of Nickel under Pressure. 389 



5. An important distinction has been drawn in paragraph 3 

 between lines which reverse and those which do not. The 

 examination of the effect o£ pressure upon the iron arc * 

 brought to light the fact that when a line was unsymme- 

 trieally reversed under pressure, its reversal was not, as a 

 rule, displaced as much as the line when not reversed, because 

 the absorption line did not coincide with the ordinate of 

 maximum emission. The symmetrically reversed lines 

 behaved normally, the reversed and bright parts being equally 

 displaced. 



As far as lines classed as unsymmetrieally reversed 

 (Class II.) are concerned, the discrepancy between spark 

 and arc is explicable on the ground that the absorption lines 

 are less unsymmetrieally disposed upon the arc lines than upon 

 the corresponding spark lines, and this view is supported by 

 a careful comparison of the appearances of these lines upon 

 the photographs which accompany the two papers under 

 discussion. It is quite obvious that the following are less 

 symmetrically reversed in the spark spectrum : — 3548*34, 

 3597-K4, 3610*G0, 3612*80, 3775-71, 3807*30, &c. But the 

 same explanation is not applicable to the symmetrically 

 reversed lines, since if the bright lines are equally displaced 

 in arc and spark, their reversals should likewise agree. That 

 they do not may be due to one of two causes : either the 

 reversals are not really symmetrical in the spark, or the 

 emission lines are displaced by different amounts in the arc 

 and spark at the same pressure. From the study of the 

 arc lines, which were seldom, if ever, truly symmetrical (on 

 which account I empWed the term " nearly symmetrical " 

 to denote lines which were nearly but not quite symme- 

 trically reversed), I suspected the former as the more pro- 

 bable explanation. 



It is with some hesitation that I question the descriptions 

 of spectrum lines recorded by an observer who has dealt 

 with the original photographs, but from the excellent plates 

 appended to Mr. Bilham's paper I should conclude that many 

 of the lines which he classifies as symmetrically reversed are 

 unsymmetrical : for example, 3510*52, 3566*55, 3619*54, 

 3624*89, 3783-67, 3831*82; and I am doubtful about all the 

 others. This means that most if not all of the lines in 

 Class I. should be included in Class II. and designated i\. 



I incline to the view that the differences between the arc 



and spark displacements (which are only pronounced in the 



case of reversed lines) are due to the differences in the 



positions of the absorption lines upon bright lines whose 



* Duffield, Phil. Trans. Roy. Soc. A. ccviii, p. Ill (1908). 



