ON THE SPECTRUM OF y CYGNL 
209 
he says,'"' “ In the ‘ Astrophysical Journal’ for June last is a list of 339 lines in the 
spectrum of the corona, photographed by Humphreys . Of these, no fewer 
than 209 do not differ from lines we have measured in the most volatile gases of the 
atmosphere, or of krypton, or xenon, by more than one unit of wave-length on 
Angstrom’s scale, a quantity within the limit of probable error.” 
It may be here pointed out that Humphreys’ list of 339 lines referred to the 
spectrum of the solar chromosphere, and not to that of the corona. The latitude 
allowed (one tenth-metre) in comparing the wave-lengths of the lines in the solar and 
terrestrial spectra is far greater than can be accepted in modern exact work, and as 
the average error of Humphreys’ wave-lengths is probably less than 0’2 tenth-metre, 
it is obvious that, until Professor Dewar can give the wave-lengths of his lines to a 
greater accuracy than that of the ]iearest tenth-metre, little weight can be attached 
to the results of his comparison. His conclusion, moreover, appears to have been 
based merely on apparent similarity of wave-lengths, without taking into account the 
relative intensities of the lines in the sjDectra compared, or of the correspondence of 
conspicuous groupings of lines, which would certainly tend to clear matters. 
The extreme limits of Humphreys’ 339 eclipse lines are, roughly speaking, 2000 
tenth-metres apart, which gives an average interval of 6 tenth-metres. In Professor 
Dewar’s three lists of gaseous lines there occur between the same limits 564 lines, 
with an average interval of 4 tenth-metres. If we assume, then, that the lines of 
each set are evenly distributed over the region involved, there will be certain to be a 
large number of lines in the two sets which agree in position within the limits of error 
allowed (one tenth-metre). 
Many lines have gaseous origins assigned to them which have been hitherto 
universally acknowledged by the various workers in the subject to be representatives 
of well-known metallic lines, and groups of lines previously given as due to some 
particular metal are sjJit up by Professor Dewar’s analysis, some members being 
ascribed to krypton, others to xenon, &c., wliile other members remain clear of his 
gaseous lines. The following table contains several groups of chromospheric lines, 
which are all included in both PIumphreys’ listf and that given in the Kensington 
eclipse publication,! and whicli have been ascribed to the same metals in the two 
records. In the comparison column, Liveing and Dewar’s gaseous lines are given 
which agree within one tenth-metre (this Ijeing the difference accepted by Professor 
Dewar in his analytical comparison) with tlie cliromospheric lines. 
‘Nature,’ vol. 66, p. 475. 
t ‘Astrophysical Journal,’ vol. 15, p. 318. 
I ‘Phil. Trans.,’ A, vol. 197, p. 208. 
VOL. COT.-A. 
2 E 
