The critical potentials of mercury, as found by Franck and Einsporn using the modified Lenard method, were studied by a further extension of this method which separated the effects due to metastable atoms and those due to true radiation. Quartz and calcite filters were interposed between the excitation system (hot-cathode and grid) and the detecting system ("photo-electric" plate and grid) of a four-electrode tube containing mercury vapor. The critical potential curves ("photo-electric" current plotted against voltage on excitation system) taken with and without these filters interposed were compared, the difference in the currents giving a measure of the relative number of metastable atoms produced at each voltage. All the breaks found by Franck and Einsporn except those at 5.76, 6.73, and 8.35 volts were found to be associated with increased production of metastable atoms. The hitherto unexplained breaks found by them at 6.04, 6.30, 7.12, 7.46, and 8.09 volts were found to be due mainly to the formation of metastable atoms. The difference between the photo-electric currents with the quartz and calcite filters interposed gave a measure of the radiation lying between $\lambda 1650$ and $\lambda 2200$, that is of the molecular bands $\lambda 2140$ and of $\lambda 1849$. The latter radiation was found to be the cause not only of the 6.7 volt break but also of that at 8.35 volts. The break at 7.73 volts is interpreted as the combined effect of radiation due to excitation at 7.69 volts and at 7.83 volts. The use of "photo-electric" plates of metals with different photo-electric characteristics gave further checks on the above conclusions.

• Received 26 June 1926

DOI:https://doi.org/10.1103/PhysRev.28.962