568 Lord Rayleigh of the bulb and connected to a pair of oil insulated condensers of 0-011 microfarads capacity each. The other sides of these condensers were connected to a high potential transformer with a 1 cm spark gap placed across. An air blast played on the spark gap.
Active Nitrogen and Law of Decay 569 to this later. In the meantime, attention has been concentrated on the containing vessel. The quality of the gas has been kept constant, and of a kind which is known to be satisfactory for the purpose. This is com mercial cylinder nitrogen which has stood over moist phosphorous, and has been dried over phosphorus pentoxide. An aspirator bottle containing the nitrogen stood ready, connected to the apparatus by a long drying tube. The volume between two taps served for admitting a standard dose of nitrogen to the bulb. When only one bulb was in use, the admission of one dose gave a pressure of 0-0035 cm. Since only small quantities of gas were taken, these were drawn from the gas which had been standing in the drying tube, and was therefore well dried.
Active Nitrogen and Law Decay 571 The sulphuric acid having been washed out, some phosphorus pent- oxide was introduced, and allowed to deliquesce. The bulb was covered by turning it around. After pouring off as far as possible, some further undeliquesced phosphorus pentoxide was added, so that there could be no suspicion of water vapour. This remained at the bottom, and only covered a small part of the area, the rest having a coating of the gummy deliquesced substance. The bulb was attached again, and a vacuum established. After 24 hours for drying the glowing gas was passed in. The value of T was 80 minutes, and further tests gave about the same result, a little better if anything than the sulphuric acid coating. This is the best result that has so far been got. The whole time that the glow remained visible after excitation was at least 5^ hours. It was still fairly conspicuous at the end of this period, and could probably have been followed for an hour or two longer. This duration is believed to be a large multiple of any obtained hitherto.
572 Lord Rayleigh There must evidently be an ideal limit to the improvement that can be made in eliminating wall effect. It seems probable that glass moistened with sulphuric or metaphosphoric acid approaches this limit, and that in this case the reaction takes place almost entirely in the gas space, without interference by the walls.
Active Nitrogen and Law of Decay 573 to the wall effect asserting itself. My own experiments were carried to considerably lower pressures, and I have regarded them as indicating that the time reaches a constant limiting value as the pressure is indefinitely diminished. The wall effect in my vessel is enormously reduced as compared with the uncoated glass bulb used by Kneser; the entire dura tion of the glow in his vessel being from 2 to 5 minutes, whereas in mine it was more than 5 hours.
574 Lord Rayleigh The neutral glasses are so selected that the density (log10 opacity) proceeds by steps of 0*10, and the glasses are numbered accordingly. These numbers are the readings of the instrument, and the result (analogous to magnitude in stellar photometry) is here called the photometric num ber. The intensity is found from it by reference to a table of anti logarithms; but it is convenient to note that a diminutive of 3 in the photometric number diminishes the intensity in the ratio 1 -995 to 1. For simplicity, and in view of the limited accuracy, I shall take this ratio as 2: 1.
Active Nitrogen and Law of Decay 575 Table I—Bulb Surface Coated with Metaphosphoric Acid Photo- Intensity Time Time in Stage metric (uranium minutes. Interval r number standard 1st 2nd 3rd Mean of minutes = 1) run run run 3 runs m. s. m. s. m. S.