On the effect of concentration on the spectra luminous gase

It is now more than forty years since Lockyer made the remarkable observation that in the spectra of electric spark discharges in mixtures of nitrogen and oxygen, the nitrogen lines were narrow and the oxygen lines broad when the oxygen was present in excess, and in the same way the oxygen lines were narrow and the nitrogen lines broad when the nitrogen was in excess. Although Lockyer put this discovery to practical use in order to make accurate measurements of the wave-lengths of the lines, it seems to have been relegated since to the numerous phenomena in spectroscopy which defy an explanation. Effects akin to this are by no means uncommon. It has long been known that the widths of spectrum lines from flames containing sodium or lithium are greatly affected by the concentration of these substances in the flame. Lord Rayleigh remarks, in connection with the behaviour of the D lines of sodium in the Bunsen flame, “Is there no distinction in kind between encounters first of two sodium atoms and secondly of one sodium atom and an atom, say, of nitrogen ? The behaviour of soda dames shows that there is. Otherwise it seems impossible to explain the great effect of relatively very small additions of soda in presence of large quantities of other gases. The phenomena suggest that the failure of the least coloured dames to give so high an interference as is calculated from Doppler’s principle may be due to encounters with other gases, but that the rapid falling off when the supply of soda is increased is due to something special. This might be of a quasi­ chemical character, e.g., to temporary associations of atoms, or again to vibrators in close proximity putting one another out of tune.” Since these words were written our knowledge of the circumstances which govern the widths of broadened spectrum lines under certain specified conditions has materially increased. Stark’s suggestion,‡ that the broadening of the lines in the spectra of condensed spark discharges is intimately connected with the resolution of the lines into components by the electric field, has been fully confirmed, and it has been shown that in the case of hydrogen and helium the broadening observed under these conditions can be accounted for satisfactorily and completely by the resolution of the lines by the electric fields of neighbouring charged particles on the radiating atoms. The electrical resolution of the lines of hydrogen and helium has been examined by a number of observers, and reliable data, both qualitative and quantitative, are available; in the case of other elements, though considerable progress has been made, our information is less complete, but it is known that for a given electric field the resolution of the lines of heavier atoms such as sodium is very small in comparison with that of the hydrogen or helium lines. It is difficult, therefore, to account for the behaviour of the lines of sodium and many other lines of heavy elements which broaden easily to an extent which seems quite out of proportion to their electrical resolution.