The ‘Proceedings of the Royal Society’ for 1860 contain a paper by Plücker, which gives an account of the first observations of the spectrum of the luminous discharge through mercury vapour at a low pressure. Plücker used a vacuum tube with mercury electrodes, and he observed and made measurements of the wave-lengths of ten lines. A few years later, working with Hittorf, he found that the mercury spectrum may be obtained more brightly when a Leyden jar and spark gap are used in parallel with the tube. Other conditions affecting the lines observed in the vacuum tube spectrum of mercury have since been recorded by various investigators; for instance, the widening of the lines with increased pressure was observed by Ciamician, and the effect of the presence of different gases in the vacuum tube on the brightness of the mercury lines was investigated by Sundell, who found that the mercury lines were visible when the tube contained hydrogen at considerable pressures, but that with oxygen or nitrogen they could only be seen when the pressure was very low. The spectrum of the light from the mercury arc was first investigated by Liveing and Dewar, and afterwards very completely by Kayser and Runge, but the first thorough investigation of the spectrum of mercury in vacuum tubes was that of Eder and Valenta, published in 1894. These observers found that the lines obtained from a vacuum tube at low pressures were much sharper than those given by the arc or spark. The number of lines obtained depended on the current density and on the temperature of the vapour. The spectrum richest in lines was obtained by having one part of the vacuum tube hot and the rest cold, so that the mercury distilled through the capillary. Using a Leyden jar they were then able to measure a great many new lines. From the wider parts of the vacuum tube they obtained a banded spectrum, but this (which was first observed by them) was seen best in the capillary when the discharge was passed without a Leyden jar. Introducing capacity into the circuit had the effect of breaking up these bands into an immense number of fine lines—the “rich line spectrum”—some 670 lines are recorded in their paper.
Read full abstract