Abstract
The generally accepted analysis of the arc spectrum of mercury into series is scattered throughout a number of papers by Paschen and a paper by Wiedmann. It shows the spectrum to contain a triplet and a singlet system of lines. Each system comprises the four main types of series—principal, diffuse, sharp and fundamental—and there are combinations and intercombinations among the systems. The scheme is partly collected by Dunz, and again by McLennan, but in neither case is the complete arrangement given, so far as it is understood at present. Such a collection presents some difficulty, especially with regard to the diffuse triplet and singlet series and their combinations. Wiedmann gives the diffuse triplet series in such a form that the three components of a triplet, with satellites, contain respectively 3, 3 and 2 lines, instead of the customary 3, 2 and 1 lines. With this classification, he draws attention to a remarkable frequency difference relation between this series and the diffuse singlet series and its combinations. But, if this arrangement is adopted, the strong series of chief lines of the first components of the triplets, as given by Dunz, is left unaccounted for. It is possible, moreover, to retain Dunz’s triplet, which has the ordinary form, and, by designating the then outstanding lines of Wiedmann’s triplet as lines of combination series, to make the aforesaid frequency-difference relation a necessary result. This arrangement appears, on the whole, to be the better of the two. It has, however, the disadvantages that it involves the splitting up of what appear to be close associated groups of lines into members of different series, and, further, that it gives combination series, 2P —
md"
, 2P —
md'
— both consisting of very strong lines—without a trace of the corresponding series, 2P —
md
, arising from the first chief line of the triplet.
Suggested Rearrangement in Quadruplets
. In the present paper a new arrangement is attempted, which overcomes these difficulties. The triplets are associated with certain singlet series, and converted into quadruplets. It will be seen from the Tables which follow that, so far as numerical relations are concerned, the evidence for the recognition of quadruplets is complete. The characters of the lines, also, in the series which have been brought together, are quite in keeping with such association. The lines of the principal triplet series are said by Wiedmann to resemble in appearance those of the series with which they are here associated so closely as to be almost indistinguishable from them. He finds the distinction, however, in the fact that the latter lines are shaded towards the violet, while the triplets are shaded towards the red. In the case of the diffuse quadruplets, the added component, consisting of three lines, contains what is usually regarded as the diffuse singlet, and a pair of combination lines, which are equally strong and diffuse. The new component of the sharp quadruplet is described by Wiedmann as being particularly sharp. It appears, therefore, that there are strong reasons for supposing that we have in mercury a system of quadruplet series.