On the other hand, the spectroscope has brought us evidence that far aloft in the atmosphere, many miles above the earth, ozone is quite abundant. Here it is supposed to be generated by two agencies--the electrical discharges of the aurora and ultra-violet radiations from the sun. The ultra-violet rays that help to produce it are prevented from reaching the earth, and astronomers are thus deprived of much interesting information they might otherwise obtain concerning the spectra of the sun and stars. However, as the present Lord Rayleigh has pointed out, we can console ourselves for this fact by reflecting that if the ozone did not shut off much of the ultra-violet light from the sun, this light would probably ruin our eyesight; or, rather, we should be put to the inconvenience of constantly wearing some sort of protective spectacles in the daytime. The high-level ozone is further interesting because of exercising a certain control over the temperature of the lower air. It is more transparent for incoming solar radiation than for outgoing earth radiation. Hence, when it is unusually abundant, it should raise the general temperature of the earth. This presumably happens when the condition of the sun is such that an unusual amount of ultra-violet radiation reaches the upper atmosphere, a fact that must be taken into consideration in any attempt to establish a relation between climatic fluctuations and the sun-spot period. The lowest part of our atmosphere is the densest because it is compressed by the weight of the air above it. Thus it happens that, although the atmosphere is at least several hundred miles in height, one-half of its mass--i. e., one-half of the quantity of matter in it, as expressed in terms of weight--lies below an altitude of about 3½ miles above sea level, while about seven-eighths lies below the ten-mile level. Above about five miles the atmosphere is too rare to support life. The highest clouds seldom occur higher than ten miles. Storms hardly ever reach that height. In short, the phenomena of life and the phenomena of weather are confined to a layer of air so shallow, in proportion to the dimensions of our globe, that on the surface of an orange it would be represented by a sheet of thin paper. The actual height of the atmosphere is not even approximately known. There are theoretical reasons for believing that even at a height of thousands of miles above the earth there are molecules of atmospheric gases still under the control of the earth’s gravity, while at such levels yet other atmospheric molecules are constantly escaping into outer space. At an altitude of fifty miles the atmosphere is less than 1/75,000 as dense as at sea level--i. e., more than seventy-five times as attenuated as the best “vacuum” obtainable with an ordinary mechanical air pump. At 300 miles it is computed to be about one two-millionth as dense as at sea level.