The series spectrum of trebly-ionised silicon (Si IV)

The investigation dealt with in the present paper forms part of a more extended inquiry into the various spectra of silicon which has been in progress during several years. The work was undertaken in the first instance in the hope pf reaching a more complete physical explanation of instance in the hope of reaching a more complete physical explanation of the stellar spectra in which the different classes of silicon lines make their appearance. Lockyer and his assistants had already shown that certain lines, which successively appeared in laboratory experiments as the energy of excitation was increased, were corresponding represented in the spectra of stars following each other in order of increasing temperature. Four groups of silicon lines were thus recognised, namely :- Group I. λλ 3905.8, 4103.2 (Arc lines). Group II. λλ3853.9, 3856.1, 3862.7, 4128.1, 4131.1, 5042, 5057 (Spark lines). Group III. λλ4552.8, 4568.0, 4574.9 Group IV. λλ4089.1, 4116.4} (vacuum tubes). Lines of Group I occur in the solar spectrum, and in stars of types G and k; those of Group II are especially marked in stars of types F and A; while those of Groups III and IV occur in successively hotter stars of types B. It seemed that the significance of the changes in the silicon spectrum might probably be better understood if the characteristics of the series including the different groups of lines could be ascertained. Bohr's theory had already given a completely satisfactory explanation. Bohr's theory had already given a completely satisfactory explanation of the change from subsequent work by the enhanced spectrum in the case of helium, and held for the enhanced (spark) spectra of the alkaline earth elements. These investigations had shown that, in passing from arc to spark lines, the series constant was changed from Rydberg's N to 4N, and the theory indicated the possibility of other types of series in which the series constant might take the still higher values, 9N, 16N, and so on. The stellar observations thus suggested silicon as a promising element in the search for these higher types of series. Stellar spectra, however, cannot be observed over a wide enough range for the investigation of series relationships, and more extended laboratory observations were therefore necessary.