Abstract
Observations were made at NorthβEast Land to determine the amount of ozone in the upper atmosphere by measuring the absorption of the ultraβviolet light in the spectrum of the Pole star. A quartz spectrograph was used and the spectra were examined by a photometer. The densities of the spectra for wave length 3,290 and 3,092 A were compared with the equal densities on an image on the plate of a source of light whose intensity varied along the image according to a known law. The constants were determined by carrying out a similar experiment at Oxford at night when the amount of ozone on the previous and following days had been measured by the standard method. Trouble was caused by the fogging of the plates caused by auroral light, but this was corrected for by a statistical method. The mean error of each determination was of the order of 30Γ10β3 cms., but the dayβtoβday fluctuation was found to be far greater than this and than the fluctuation observed in lower latitudes. There was no apparent connection between the amount of aurorΓ¦ and the amount of ozone. The results show that the autumn fall in the amount of ozone, observed at all latitudes is continued through the winter and is followed by a sudden rise in spring.