374 
DR. T. R. MERTON AND MR. S. BARRATT 
records in the green regions of the spectrum, in which the sensibility of the eye is a 
maximum and that of most panchromatic plates a minimum. Frost ( loc. cit.) and 
Ames (loc. cit.) have published short lists of some of the stronger lines, but the most 
comprehensive and accurate table is due to Watson (loc. cit.), who has recorded most 
of the lines in the red and yellow, and in the ultra-violet, but has not included a con¬ 
siderable number of lines in the green regions of the spectrum. Watson found no 
lines of wave-length greater than the Balmer line Ha. Porlezza (loc. cit.) and Porlezza 
and Norzi (loc. cit.) have published tables which supplement those of Watson, and 
Croze (loc. cit.) has measured lines in the infra-red down to 78000A. 
The measurements included in the present investigation extend from Ha to the limits 
imposed in the ultra-violet by the thin glass wall of the discharge-tube, the shortest 
wave-length recorded being 73375 A, but the continuous spectrum of Hydrogen could 
be traced on the plates to wave-lengths shorter than 73000 A, and as there were no 
indications of lines superposed on the background in this region, we did not resort to 
the use of vacuum-tubes provided with quartz windows. It is intended ultimately to 
extend the measurements into the infra-red. We have used an Anderson concave 
grating ruled with 20,000 lines to the inch, and having a radius of curvature of 120 cm., 
which gave a dispersion of very nearly 10 A per millimetre. The mounting was of the 
type described by Eagle ( £ Astrophys. Journ.,’ 31, p. 120, 1910), which involves three 
adjustments in focussing, of which, two determine the angles made respectively by the 
grating and by the photographic plate with the incident light, whilst the third is used 
to vary the distance between the grating and the plate. The necessary adjustments for 
different regions of the spectrum were found from experimentally prepared tables. 
Plates of especially thin glass were used and were bent to the appropriate curvature in 
the plate-liolder. 
The regions from 76560 to 75400 were photographed on Wratten and Wainwright 
Panchromatic plates, from 75400 to 74860 on Marion’s Iso-Record plates which are 
specially sensitive to this region, and from 74860 to the ultra-violet on Ilford Ordinary 
plates. The vacuum tubes were used end-on, the light from the capillary being focussed 
upon the slit of the spectrograph by means of a quartz lens of about 30 cm. focal length. 
The exposures required to bring up the faintest lines which were measured were five 
hours with the Panchromatic plates and three and a half hours with the Iso-Record and 
Ilford plates. The International Secondary standards were used as a comparison 
spectrum, the source of light being a Pfund ( £ Astrophys. Journ.,’ 27, p. 296, 1908) arc 
burning with a current of about 3 amps, at 100 volts. The comparison spectrum was 
limited by a movable stop in the spectrograph to a narrow strip running across the 
middle of the Hydrogen spectrum. It was found impossible to ensure the absence of 
very small shifts between the Hydrogen and the comparison spectra, which were photo¬ 
graphed consecutively on the same plate, and to eliminate errors, due to these shifts, 
from the measurements the following procedure was adopted. A tube containing 
Helium and Hydrogen was substituted for the tube containing pure Hydrogen and a 
