THE Electrical World, in referring to a contribution by Mr. W. W. Coblentz to a recent issue of the Physical Review on the probable temperature of the sun and moon, says: It appears that the radiation power of the sun, at the surface of the earth, amounts to about 1.75 kilowatts per square meter of perpendicularly exposed earth-surface. Of this, the books tell us about one-quarter is absorbed in the air, when the sun is at its zenith, or getting in his best work; so that what reaches the earth when the sun is overhead is, say, about 1.3 kilowatts per square meter. In the temperate zones, where the sun is never vertically overhead, the layer of air passed through by the sun's rays is thicker and the absorption consequently greater, especially in early morning and late evening, so that a square meter of surface kept facing the sun all day long during a clear summer day might only receive an average radiation power of about 0.5 kilowatt. Of course, the square meter would reflect away a large proportion of this power, if its surface were of polished 1NTERIOR OF MAGNETIC OBSERVATORY- DECLINOMETER AND THEODOLITE. metal, and even a dull black surface, like that of plumbago, would dissipate convectively the heat which it received, so that it is very hard to catch and utilize this radiated solar power. Nevertheless, if we could employ this power practically and conveniently, we should obtain an immense benefit. Thus, allowing that the noon-day solar power on a bright day -was 1 kilowatt per square meter of perpendicularly exposed surface, we should only have to expose a surface of 10 meters square in order to receive 100 kilowatts; and if an efficiency of 50 per cent were imagined in the apparatus, we would be able to develop 50 kilowatts during the brightest part of the day from a disk about 37 feet in diameter. The only solar engine which has yet been made successful is the waterfall. A fraction of the solar radiation energy reaching the surface of the earth is expended in converting surface ocean-water into steam or water vapor and in raising that steam to an elevation among the clouds. Part of this energy is released in rainfall, and only an insignificantly small fraction of the rainfall occurs on elevated land in such a manner that a waterfall can be made available. There is at least one other type of solar engine possible, and that is a surface of chemical substance exposed to solar radiation and capable of being chemically transformed to a stable substance which will subsequently give up its energy for consumption. A grass meadow supporting horses is a crude form of such a machine. A small fraction of the incident solar energy is usefully absorbed by the chlorophyl in the grasses, permitting them to build up a hydrocarbon structure ' from an environment of gaseous water and carbon-dioxide. The horses consume and assimilate the grass, and each is capable of delivering a few kilowatt-hours a day of solar energy—an infinitesimal fraction of the total solar energy incident on the meadow. It might be possible to find a chemical substance much superior to chlorophyl as a recipient or storage material, and capable of releasing its energy in an electrical way. The paper shows that the surface temperature of the sun works out, about, 5,980deg. absolute or 5,707 deg. C., each square meter of solar surface liberating apparently 67,600 kilowatts or not far from 90,000 horse-power. The effective temperature of the moon on the side facing the sun ap pears to be about 82 deg. C. This shows how small a share of incident radiation energy a reflector can claim as commission for its duty. The moon is bupposed to have little else to do, from a human utility standpoint, than to reflect radiation. She constantly receives a large total amount of radiation power, but is not able to raise her surface temperature thereby beyond about 100 deg. C.