A previously unrecorded band, stretching from about λ 4100 to λ 4200, was observed by Prof. A. Fowler on a photograph of the vacuum-arc spectrum of silicon taken with the first order of a 10-foot grating in 1919, and was considered to be most probably due to silicon hydride. In 1921 Prof. Wood, in the course of his work on the hydrogen spectrum, observed a band of similar appearance, which he also considered to be due to a compound of silicon (from the glass walls of the discharge tube) and hydrogen. He gave a list of the wave-lengths of the band lines, and these agree well with the prominent lines on Prof. Fowler’s plate, although the latter contains many more lines. In the course of the present work it was found that the band observed by Prof. Fowler was in reality a mixture of the spectrum of silicon hydride and that of silicon nitride. The band of Prof. Wood’s plate includes, however, only lines now attributed to the compound SiH.
Experimental
. The conditions most favourable for the production of the band were first investigated with the aid of a spectrograph of very wide aperture and low dispersion (about 40 A. per millimetre at λ 4100). Photographs of the spectra of the silicon arc
in vacuo
, in water vapour, in hydrogen at atmospheric pressure and at very low pressure, were taken, but these sources all appear to be much inferior to the arc in hydrogen at pressures between about 3 and 9 cm. of mercury. Using the latter source photographs on a larger scale were then taken with the 10-foot grating spectrograph. Using the first order (5.5 A. per millimetre) a strong plate was obtained with an exposure of about ¼ hour. The current was approximately 6 amperes at 110 volts, and the pressure of hydrogen about 4 cm. of mercury.
The spectrum of silicon hydride
