Using the method of Beams and others, the time intervals between the appearance in spark discharge spectra of about fifty lines of sodium, lithium, calcium, barium, strontium, magnesium, cadmium, zinc, and nitrogen were measured for the purpose of determining if there was any relation between these times and the nature of the lines or the positions of the elements emitting them in the periodic table. The time intervals for different lines varried from zero to about 2×10-8 second. Most of the lines seemed to be emitted at the very beginning of the discharge, before the potential across the spark gap had dropped appreciably, There did not appear to be any definite relation between the time intervals and the nature or origin of the lines. With one exception, the members of multiplets which could be resolved appeared simultaneously.
A variation of the time intervals with the amount of inductance in the spark gap circuit was observed and an explanation of this is given; the times were independent of the other electrical characteristics of the circuits. In a qualitative theory of the beginning of spark discharge spectra, it is suggested that the time of appearance of any line depends not only on the “life” of the excited atoms, during which there is no emission, but also on the minimum current in the discharge required to excite the line with a visible intensity.
GORDON L. LOCHER, "THE TIME INTERVALS BETWEEN THE APPEARANCE OF SPECTRAL LINES IN THE SPECTRA OF ALKALI AND ALKALINE-EARTH METALS," J. Opt. Soc. Am. 17, 91-99 (1928)