IN a communication on the optical excitation of mercury vapor, which was published in the Philosophical Magazine for September, 1927, I drew attention to the circumstance that certain intensity ratios of some of the mercury lines could be explained only by assuming that the intensities of some of the emitted lines increased with the square of the intensity of the exciting light, while others increased with the cube . The reasonableness of this assumption was explained as follows :
The vapor is excited by the total radiation of a watercooled mercury arc . By the absorption of the radiation of wave-length 2,536 the normal mercury atoms are brought into the excited state with electrons on the 2p2 energy level (or orbit) . These atoms are in a condition to absorb the 4,358 radiation, by which process they are brought into a higher excited state, the electrons being raised to the 2s level . From this level they may return to the lower levels 2p,, 2P2, 2P 3 , with the emission of the lines 5,461, 4,358 and 4,046 . The atoms which were responsible for the emission of the green line 5,461 are now in the meta-stable condition with electrons on 2p i , and can absorb the radiations 3,650, by which they are brought to a still higher excited state with electrons on the d orbit. From this level the only transition which can take place is a return of the electrons to 2p1, with emission of the line 3,650, and this is the only way in which this line can be optically excited . These transitions are shown in