330 METEOROLOGY. 



The following cases under tlie general formula may liere be specified : 



First, at the time of the equinoxes, the sun's daily intensity for all 

 places on the earth is proj^ordonal to the cosine of the latitude. As the 

 equinoxes in March and September lie intermediate between the ex- 

 tremes or maxima of heat and light in summer, and their minima in 

 winter, the presumption naturally arises that the same expression will 

 approximate to the mean annual intensity. The coincidence is ac- 

 cordingly worthy of note, that the best empirical expression now known 

 for the annual temperature in degrees Fahrenheit, given by Sir David 

 Brewster, in the Edinburgh Philosophical Transactions, Vol. IX, is 

 81°. 5 cos L, being also proportional to the cosine of the latitude. It 

 is remarkai)le that Fahrenheit, in 1*720, should have adjusted his 

 scale of temperature to such value, that this formula applies, without 

 the addition of a constant term. 



Secondly, for all places on the equator the latitude is ; and the 

 sun rises and sets at six, the year round, exclusive of refraction. Con- 

 sequently the sun's diurnal intensity varies slowly from one day to 

 another, being proportional to the cosine of the meridian declination of 

 the sun. 



Thirdly, at the north pole, the sun rises only at the vernal equinox 

 in March, and continues wholly above the horizon, till it sets at the 

 autumnal equinox. Thus, to either pole, the sun rises but once, and 

 sets but once in the whole year, giving nearly six months day, and 

 six months night. Now suppose the six months day to be divided 

 into equal portions of twenty-four hours each, then the intensity during 

 twenty-fonr hours of p)olar day is proportional to the sine of the declina- 

 tion at the middle of the day. 



Fourthly, at the summer solstice, when the intensity on the pole is 

 a maximum, the ratio becomes as 1 to 1.25 ; or the polar intensity is 

 one-fourth part greater than on the equator. The difference evidently 

 arises from the fact that daylight in the one place lasts but twelve 

 hours out of twenty-four, while at the pole the sun shines on through 

 the whole twenty-four hours. 



It were interesting to find when this polar excess begins and ends, 

 which is ascertained to be on May 10th, and again on August 3d. 

 Therefore, during this long iJiterval of eighty-five days, comp)rehending 

 nearly the lohole season of summer, the sun's vertical intensity over the 

 north pole is greater than upjon the equator. To this subject we shall 

 again recur in a subsequent section. 



Fifthly, having glanced at these particular cases, let a more com- 

 plete survey be made for the northern hemisphere. A.nd the same 

 will equally apply to the southern hemisphere, allowing for the rever- 

 sal of the seasons and change of the sun's distance. 



The subjoined table has been computed for intervals of fifteen days, 

 and expresses the results in units of intensity. The choice of a unit. 

 being entirely arbitrary, the intensity of a day on the equator at the 

 time of the vernal equinox is here assumed to be 81.5, and other 

 values are expressed in that proportion. In the last three columns 

 for the frigid zone, the braces include values for the days when the 

 sun shines through the whole twenty-four hours ; the blank spaces 

 indicate periods of constant night. 



