The intensity of their spectral lines is one prominent indication of the presence of elements in the sun, and the number of coincidences of spectral lines is another. Iron, nickel, calcium, manganese, sodium, cobalt, and carbon are among the elements most strongly identified. A few of the rarer terrestrial elements are of doubtful existence in the sun, and a very few, as gold, bismuth, antimony, and sulphur are not found there, and the existence of oxygen in the sun is regarded by some experts as doubtful. But if the whole earth were vaporized by heat, probably its spectrum would resemble that of the sun very closely. What are the effects of the sun, and sun spots in particular, on our weather? Is the influence of their periodicity potent or negligible? If we investigate conditions pertaining to terrestrial magnetism, as fluctuations of the magnetic needle, and the frequency of auroræ, there is no occasion for doubt of the sun's direct influence, although we are not able to say just how that influence becomes potent. If, however, we look into questions of temperature, barometric pressure, rainfall, cyclones, crops, and consequent financial conditions, we find fully as much evidence against solar influence as for it. The slight variations of the sun's light and heat due to the presence or absence of sun spots can scarcely be sensible, and much longer periods of closer observation are necessary before such questions can be finally decided. The slighter such influences are, if they actually exist, and the more veiled they are by other influences more or less powerful, the more difficult it is to discover their effects with certainty. The importance of solar radiation in the prediction of terrestrial weather has long been recognized, but until very recently no practical application has been made. The Smithsonian Astrophysical Observatory at Washington, under the direction of Dr. Abbot, has for many years carried on at a number of stations a series of determinations of the constant of solar radiation by the spectro-bolometric method originated by Langley. A new station in Calama, Chile, has recently been inaugurated, at which the solar constant is worked out each day, and telegraphed to the Argentine weather service, where it is employed in forecasting for the day. Abbot's new method of solar constant determination is based on the fact that atmospheric transparency varies oppositely to the variations of brightness of the sky. Increase of haziness presents more reflecting surface to scatter the solar rays indirectly to the earth. Of course it presents also additional surface to obstruct the direct rays from the sun. By measuring the brightness of the sky near the sun, it becomes possible to infer the coefficients of atmospheric transmission at all wave lengths. The direct observations and the complete deduction of the solar constant for the day can all be completed within two or three hours.