Page:The American Cyclopædia (1879) Volume XI.djvu/446

434 METEOROLOGY Attempts have been made to demonstrate other periods, of which perhaps the only one whose existence is plausible appears to agree with the time of rotation of the sun upon its axis. We are therefore justified in considering the intrinsic radiation from the sun as very approx- imately constant, and the diurnal and annual variations of terrestrial temperature depend upon the position of the station on the earth's surface, and the position of the earth's axis of rotation in reference to the annual orbit de- scribed by the earth about the sun. The quan- tity of heat received by any surface varies directly as the time of exposure and as the sine of the sun's altitude, and inversely as the square of the sun's distance. According to Lambert (1770) and Meech (1855), the sun's daily intensity is proportional to the cosine of the latitude. At either pole the intensity in midsummer is one fourth greater than on the equator ; this arises from the fact that daylight on the equator lasts but 12 hours, while at the pole the sun shines throughout the whole 24 hours. In general, from May 10 to Aug. 3 the sun's vertical intensity over the north pole is greater than upon the equator. In the tem- perate zone the temperature of the air attains its maximum about one month after the maxi- mum of the sun's intensity ; in this interval therefore the earth must receive during the day more heat than it loses by radiation at night. The average annual intensity upon the whole earth's surface from pole to pole is 299 ther- mal days, the intensity of each of which units equals that of the mean equatorial day. The annual intensity of solar rays during 100,000 years, past or future, can never vary (owing to the varying eccentricity of the earth's orbit) more than the equivalent of five hours of average sunshine in a year. These conclusions refer to the whole earth's surface collectively. On the other hand, the annual intensity at the different latitudes oh the earth varies much more considerably, and the extreme values of diurnal intensity may vary by as much as one ninth of the present value. It seems therefore that under the existing conditions of physical astronomy the intensity of solar heat upon the earth can never have been materially different from its present value. In these rules we have considered only the heat received at the outer surface of the atmosphere from the sun, neg- lecting the absorption of heat by the earth's atmosphere and the radiation of heat back into space, two circumstances that materially affect actual temperatures. The absorption of heat in its passage through the atmosphere is directly found approximately by observations made with the pyrheliometer of Pouillet (Pog- gendorff s Annalen, xlv.), or the actinometer of Herschel (1825), which instruments replace the ruder contrivances of earlier days, such as Leslie's photometer (1797), and De Saussure's helio-thermometer (1787). The only self-re- cording apparatus that indicates the power of the direct rays of the sun at present in use is the so-called black-bulb-in-vacuum or solar- radiation thermometer of Negretti and Zam- bra. Observations with Pouillet's pyrheliom- eter show that the absorption of solar heat by the atmosphere follows sensibly the same law as the absorption of solar light, and amounts for the temperate zone to from 20 to 40 per cent, when the rays penetrate vertically down- ward. According to the investigations of ,Melloni, Tyndall, Magnus, and others, aqueous vapor is almost opaque to the invisible heat rays belonging to the red end of the spectrum, and accordingly an increase in the absorp- tion of the direct solar rays occurs where an in- creased amount of moisture is present in the air. The absorption of solar heat by the material composing the earth's surface varies of course with every change in the constitution or mo- lecular condition of the latter. Dry and sandy or rocky soils become heated to a higher tem- perature than the moister portions of the earth, while the ocean experiences the least varia- tion of temperature. But perhaps the most important property of the earth's surface con- sists in this, that the rays which are not ab- sorbed by it, and which are consequently ra- diated back through the atmosphere, have been degraded to the red end of the spectrum ; these are therefore very largely absorbed by the aqueous vapor in the lowest portions of the atmosphere, from 40 to 90 per cent., ac- cording to the dryness of the air, being re- tained within one or two miles of the earth's surface. As a consequence of this absorp- tion, the temperature of the air at the surface rises most rapidly during the day at places covered by a layer of clear moist air, even though such layer be at a considerable altitude above the station, without reaching down to it; and on the other hand, the temperature diminishes most rapidly at night when a clear atmosphere, holding but little moisture, exists over the station. The conduction and convec- tion into the interior of the earth and ocean of the solar heat that falls upon its surface, produces such a storing up of heat as to sensi- bly ameliorate the sudden changes that would otherwise occur, and to delay the periodical daily and annual maxima and minima of atmos- pheric temperature. The temperature of the soil has been measured by means of thermom- eters whose bulbs are, as was first suggested by Quetelet, permanently imbedded therein ; and the general laws governing the distribution of temperature in the interior of the earth were mathematically investigated by Fourier (1812) and Poisson (1835). The range of the varia- tions of temperature diminishes rapidly as we descend into the soil, forming an inverse geo- metrical series when the depths form an arith- metical series; again, the durations of corre- sponding periodical changes in temperature at and below the surface of the ground increase at any depth in proportion to the square root of the durations of the surface periods. Daily variations of temperature are perceptible at a