In terrestrial experiments it is not possible to leave an atom undisturbed for more than 1/1000 of a second. Within this time it will collide either with other atoms of gas or with the walls of the vacuum tube. All atoms which find themselves in a metastable state are likely to be released within 1/1000 of a second, and they have no occasion to use the emergency exit. In the rarefied material round the sun -- the chromosphere and corona -- collisions are rare; however the strong solar radiation excites the atoms many thousands of times per second, so that they are quickly released from metastability by upward passages. But in the solitude of a nebula the atom can wander a year or more without colliding with anything, and the light traversing the nebula is so feeble that excitation occurs only about once a century. An atom which (after its last excitation) has fallen down into a metastable state hesitates a long while; but nothing intervenes to release it, so at last it takes the forbidden passage. Thus in the tranquillity of a nebula we observe light which is not emitted in the commotion of a vacuum tube.

It must not be overlooked that the enormous extension of the nebulae has a great deal to do with the apparition of these forbidden lines. We do not know how long the atom waits in a metastable condition -- a minute, a month, a century. There is no way of hurrying the atom; if we do not allow it this rest it will not emit the forbidden line. Nebulium radiation must therefore be extremely weak compared, for example, with the calcium light in the sun's chromosphere where each atom does its job 20,000 times a second (p. 72). The great number of atoms in a nebula makes up for the laziness of the individual atoms. If a nebula contains an amount of oxygen equal to the sun's mass -- a reasonable estimate for diffuse nebulae -- and if each oxygen atom is stimulated to emit nebulium light once a century, the total brightness of the nebulium light will be 100 times the brightness of the sun.

This identification of nebulium is confirmed by the fact that the other conspicuous unknown lines in the spectra of nebulae have been identified with forbidden lines. Some belong to singly ionized oxygen, the others to singly ionized nitrogen. A mixture of nitrogen and oxygen has a more familiar name. Once again we have been beguiled by that arch-humorist Nature. She has set in the heavens cloudy forms gleaming with a 'light that never was on sea or land', and we have imagined and even named strange elements composing them. The shining fluid that so long has baffled us is -- Air.