Ultraviolet-filter transmission measurements are made, using a newly devised p ortable precision ultraviolet meter (Research Paper RP647) and 4 glass filters, giving the spedal quality and total intensity in the band of ultraviolet solar radiation extending from about 2900 to 3500 A. At the high-altitude station (Flagstaff, Ariz.) data were obtained on solar ult.raviolet intensities under various conditions; high and low humidity, large and small air m asses, clear and sm oky skies. The data obtained at sea-level st ations (Washington, D. C., and San Juan, P. R.) a re of interest in connection with the question of ultraviolet intensities in the tropics as comp ared with similar stations at higher latitudes.Data are given on the amount of ultraviolet transmitted through light fog, and on the a m ount r eflected by the sky and by snow; also the effect of altitude and latitude on the sola r ultraviolet intensity at the earth's surface.The observations show that, while a light fog greatly reduces t he intensity (when the sun appeared as a distinctly outline d white disk, the intensity was reduced to one-tenth or less) the spectral quality was practically unchanged, owing to the relatively nonselective character of the transmission through water vapor. Similarly, the ultra violet refl ect ed from snow is extraordinarily high, owing to the fact that the ice crystals, within which the reflection occurs, are highly transparent to ultraviolet radiat ion.The absorption of ultraviolet is caused by atmospheric polution n ear t he earth's surface, and by ozone in the stratosphere . An increase of 40 to 50 percent in intensity in the band of wave length between 2900 and 3130 A, is observed in rising 2 km (7,000 ft) above sea level. A much greater increase in intensity may be expected at elevations of 15 to 25 km (10 t o 15 miles). However, such heights are not practicable for large-scale biological tests.The measurements at the Flagstaff station show a greater atmospheric transparency (less ozone) in the afternoon than in the forenoon ; and a greater transparency in the autumn than in the spring. This is in agreement with the work of Dobson and others, using other methods of observation.For the same solar altitude (air mass traversed by the rays) the ultraviolet solar intensities in the tropics (at San Juan) were somewhat higher than at a midlatitude, sea-level station (Washington) that is free from local air pollution . This is in agreement with other observations showing that, in the tropics, for the same solar altitude, the amount of atmospheric ozone is less, and consequently the ultraviolet intensities should be somewhat hi gher than in higher latitudes.However, this small difference in intensity, for the same solar altitude, does not appear to be sufficient to produce marked differences in biological effects.