Spectral Energy-Curve of the Sun in the Ultraviolet.

i. Appa~atus.-The Mount Wilson observations were made with the double quartz monochromator (slightly modified) that was used in the Tucson work. Light was sup- plied by an aluminized siderostat mirror. Both lens-projection and mirror-projection systems were used. 2. Measurements-The measurements were made during the years 1934, 1937, and 1939, with the thermopile, by the method of constant dispersion. The results, reduced to no atmosphere, yield spectial energy-curves for integrated sunlight and for light from the center of the disk which verify the principal features obtained at Tucson, namely, a sudden drop of 48 per cent in intensity from X 0.40 to X 0.38 j~, a nearly constant in- tensity from X 0.38 to X 0.325 ~, and a nearly linear fall of intensity to a very low value at X 0.292 ji. 3. Atmospheric transmission coefficients-The means of determinations on 35 days, which include both integrated light and light from the center of the disk, do not differ more than 2 per cent from those of the Smithsonian observers until the ozone band is reached. Corrections due to the fall of energy within the second slit have been applied to the results for X 0.295 and X 0.292 ~. 4. Atmospheric ozone content-A plot of atmospheric transmission against wave length shows that the ozone content is about 0.10 cm at N.T.P., only slightly more than half that found at Tucson. 5. Energy between the lines-To eliminate as far as possible the effects of Fraunhofer line absorption, energy-curves were obtained with a 2 i-foot concave-grating monochro- mator and photoelectric cell over 100 A units of the spectrum. The areas below contour lines drawn over the highest ordinates, divided by the areas below the record of the line spectrum, furnish coefficients by which the thermoelectric energy-curves may be re- duced to an energy-curve for the continuous spectrum. Although this procedure smooths out the ultraviolet energy-curve to some extent, most of the major depressions remaiii, and the curve in the ultraviolet is still far removed from that of a black body. 6. Band spectrum of ozone-An earlier laboratory investigation of the ozone spec- trum is reported. A calibration of the plates shows that the line intensities are so low that they will not be distinguishable in the solar spectrum and cannot alone account for depressions found in the photoelectric energy-curves. 7. Line-group intensities-From the photoelectric records of the 2 i-foot monochro- mator the line-group intensities for the regions XX 3200-3300 A and XX 3900-4000 A are given. A comparison with Mulders' calibration of Rowhnd intensities in the latter region shows an average ratio of o.8i, with considerable scattering