Revision of AlI Terms

Abstract

IN the course of a study of the proper co-ordination of the AlH spectral terms to the Al + H atomic combinations1, some doubts arose as to the correctness of Lansing's analysis2 of the Al I lines in the region 3050–3070 A. Spectrograms of an interrupted aluminium arc were taken with the large concave grating of this Institute. The measurement of the seven lines (three of which were attributed by Lansing, on the basis of wrong measurements, to an sp2 2D†s2p 2P transition) revealed, first, that no one wave-number difference is equal to the separation of the s2p 2P ground state of Al, and secondly, that all the seven observable lines can be explained at once by a 3s3p4s 4P†3s3p2 4P transition. This explanation is supported by the analogous term-combinations in the isoelectronic sequence Al I, Si II, etc.3; the separations of the multiplet components and relative positions of the two 4P terms of Al can be calculated to a rather good approximation from the positions and separations of the related terms. Preliminary results of Zeeman effect experiments on the strongest of the seven lines involved also confirm the 4P†4P explanation.