# Page:Popular Science Monthly Volume 80.djvu/528

As mentioned above, we have but few data from Tucson and, therefore, at present we can scarcely make a comparison, but doubtless when a longer series of observation is at hand we shall discover that there are many days in Tucson showing a maximum as high or even higher than Manila, and an average of about the same. The fact is also observed that two days apparently equally clear will show marked differences between each other during corresponding hours, so that the proportion of the rays reaching the earth and lying between 550 ${\displaystyle \mu \mu }$ and 291 ${\displaystyle \mu \mu }$ varies from day to day. A comparison of the total effect of these rays with the measurements obtained by the black bulb thermometer also demonstrates that the two are not functions of each other. Of course it is clear that there must be a certain relationship because, obviously, on clear days both black bulb readings and photocatalytic measurements will be high, but they need not necessarily be high in the same proportion. The sun's rays which lie in the portion of the spectrum under discussion, on reaching the atmosphere, suffer molecular scattering, refraction and other changes which modify the proportion of direct sunlight that reaches the earth's surface, and, of course, these changes vary with the condition of the atmosphere. Nevertheless, it is very interesting to observe that even in the tropics the shortest wave-lengths appearing in the spectrum of the sunlight are very close to 291 ${\displaystyle \mu \mu }$ and that no shorter rays reach the surface of the earth. The same observation was made in northern latitudes, so that it can confidently be stated that the range of the spectrum everywhere is the same, the difference, if any, being in the intensity of the light.
The above is a very brief summary of the results so far obtained in a study of photocatalytic reaction brought about between the rays lying between 550 and 291 ${\displaystyle \mu \mu }$, and in view of the results the more extensive remainder of the spectrum, which extends upward from the point mentioned into the red and infra-red and which would include the heat rays, must be considered. It is self-evident that comparative measurements in this field involve much greater difficulties, as, at present, no photocatalytic reaction is available. The best means at hand is by a comparison of a series of readings with the Angström pyrheliometer of