Arc and Spark Spectra of Nickel under Pressure. 

 Table III. 



The Displacements of Iron Lines when Pure and 

 when Diluted. 



391 



Wave-length 



(Iron). 



Nickel Spark 

 (Bilham). 



10 atms. 



Nicke 

 (Duff 



10 atms. 



1 Arc 

 ield). 



Higher 

 Pressures. 



19 



Iron Arc 

 (Duffield). 



10 atms. 

 2-5 



4045-98 



36 



4-0 



4063-76 



3-2 



30 



20 



2-5 



4071-91 



2-3 



34 



2-0 



2-3 



4383-72 



4-2 



50 



2-1 



2-0 



Means 



33 



3-8 



2-0 



2-3 



Italics indicate that the line is reversed. 



The mean value for the arc is rather higher than that for 

 the spark, but the reading for the last line is probably too 

 high. The close agreement for the bright lines in the arc 

 and spark at 10 atmospheres is thus again emphasized. The 

 fact that the diluted iron gives higher values than pure iron 

 is curious, as a density effect, if due to the interaction of 

 similar centres of vibration, would act in the opposite -way. 

 We note, however, that there is good agreement between 

 the values obtained with pure material (reversed lines) and 

 those derived from a trace of it in the nickel arc under 

 higher pressures. As the arc reversals are approximately 

 symmetrical, the difference between the pure and diluted 

 iron readings is probably due to the nature of the nickel 

 arc at 10 atmospheres, which is such that there is some 

 radical difference in the temperature and density gradient 

 of the iron vapour in the iron and nickel arcs at this 

 pressure. 



Note upon the Influence of Density and Temperature 

 Gradients upon the Displacement of Spectrum Lines. 



The difference in the displacements of the reversed parts of 

 lines in the arc and the spark (which occasions an increased 

 dissymmetry in the latter) is due to the fact that the density 

 and temperature gradients are different in the two types 

 of discharge. I incline to the view that these gradients 

 affect the positions of the bright lines also, to a small extent, 

 but for two reasons their influence is greater upon reversals. 

 First, because if we follow the different layers of the source 



