252 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 13, No. 12 
Structure of Multiplet 3 
18776 .96—384 .32—19161.28 
240.18 240.16 
18536 .78—384 .34—18921 .12—272.53—19193 .65 
199.55 199.50 
18721 .57—272.58—18994.15—175 .26—19169 .41 
139.70 139.70 
18854 .45—175 .26—19029 .71—86 .01—19115.72 
71.07 
18958 .64 
King’ gives a table of twenty-one lines which appear in the fur- 
nace spectrum of iron at a temperature of 1400°C. Twelve of the 
stronger of these lines give multiplet No. 1. 
Multiplet No. 2 is novel in that it has the same difference repeated 
three times in two instances. It will be noted that the lines involved 
_ are all reversed lines and of nearly the same temperature class. This 
anomaly occurs also in the multiplets given as Nos. 9 and 10. 
Multiplet No. 3 is interesting in that the differences 240.17, 199.52, 
139.70, and 71.07 are here related to the frequency so that in a given 
triplet, the larger difference occurs between the greater frequencies, 
while in multiplet No. 1, with the same differences, the greater differ- 
ence occurs between the smaller frequencies involved in a given 
triplet. 
The above-mentioned multiplets may be regarded as typical and 
the others are presented in Table 1 without individual comment at 
this time. It will be noted that there are altogether thirteen sets 
of separations involved in the twenty multiplets, one set recurring 
in nine multiplets. The total number of lines in Table 1 is 212. 
These include 102 of the 134 lines described by Burns as reversed in 
the arc and most of the lines of temperature classes I and II in King’s 
furnace spectra are here represented. The correlation of these multi- 
plets with temperature classification is seen to be fairly satisfactory, 
but inspection of the Zeeman patterns show that these data are 
homogeneous for some groups and quite discordant for others. 
Some interesting relations have been found to exist between the 
different multiplets but a discussion involving a physical interpretation 
of these regularities is deferred until the spectrum has been more 
fully analyzed. A systematic analysis of all the available data on 
both are and spark spectra of iron is in progress and the complete 
results of this investigation will appear later as a scientific paper of 
the Bureau of Standards. 
10 Astrophysical Jour. 45: 370. 1922. 
