The visual spectrum of the afterglow accompanying the reversion of active nitrogen to ordinary nitrogen was investigated by Prof. A. Fowler and myself, and it was shown that the remarkable green, yellow and red bands in the spectrum of this glow, originally observed by Lewis, were in reality a selection of the first positive bands of nitrogen. We were only able to make this identification when we had examined the detailed structure of the bands on a fairly large scale spectrogram. Owing to the change of relative intensities the general aspect of the spectrum is so totally different from that of the positive column of a vacuum discharge, that it fails to suggest any connection. This is an unusual instance in spectroscopic experience. It is not often that the identity of a spectrum is totally disguised by changes in relative intensity. In 1912 I made experiments in search of similar afterglows in other gases than nitrogen. The results were, for the most part, negative, but in the case of helium a curious observation was made. The gas was contained in a vessel as shown in fig. 1, a powerful jar discharge passing between the electrodes. The gas diffused out of the region of discharge into the bulbous annexe, where it displayed the afterglow, if any. On introducing helium containing, as it happened, a little nitrogen, an orange-coloured afterglow was observed, which when examined with a small spectroscope appeared to be the same as that given by nitrogen. The redder colour was explained by a shift of relative intensity of the bands towards the red end of the spectrum. No measurement of the position of the bands was made. On absorbing the nitrogen with cooled charcoal, the afterglow became still redder, and finally disappeared. Pure helium gave no afterglow.