38 INTENSITY OP SUN'S HEAT AND LIGHT. 



difference is in the Polar regions, where the secular change of annual intensity is more 

 than four times greater than on the Equator ; in its annual amount, the Polar cold 

 is now very slowly increasing from century to century, which effect must continue 

 so long as the obliquity of the ecliptic is diminishing. And thus, so far as relates 

 to a decreased annual intensity, the celebrated " North-west passage" through the 

 Arctic sea will be even more difficult in years to come than in the present age. 



Having now considered the secular changes of annual intensity upon the earth 

 and its different latitudes, let us next examine the secular changes of intensity in 

 relation to the Northern and Southern hemispheres. The earth is now nearest the 

 sun in winter of the northern hemisphere on January 1st, and farthest from the 

 sun in summer, on July 4th. This collocation of times and distances has the advan- 

 tage of rendering the extreme of summer cooler, and of winter, north of the equator, 

 warmer than it would be at a mean distance from the sun. But south of the 

 equator, on the contrary, it exaggerates the extremes by rendering the summer hotter 

 and the winter colder. Before estimating this difference, we may observe that the 

 perigee advances in longitude IP'. 8 annually; by which the instant when the earth 

 is nearest the sun, will date about five minutes in time later every year. The time 

 of perihelion which now falls in January, will at length occur in February, and 

 ultimately return to the southern hemisphere the advantage which we now possess. 

 Indeed, it is remarkable that the perigee must have coincided with the autumnal 

 equinox about 4,000 B. C, which is near the time that chronology assigns for the 

 first residence of man upon the earth. 



For ascertaining the difference of intensity, we know that the sun's declination 

 goes through a nearly regular cycle of values in a year. The formula cos H = — tan 

 L tan D then shows that the length of the day in the southern hemisphere is the 

 same as in the northern hemisphere about six months earlier. Recurring to formula 

 (18), it appears that the difference of intensities will then depend chiefly on the 

 values of A 2 . Now, for the northern winter on January 1st, A 2 is proportional to 



1 . . 1 



; for winter in the southern hemisphere, July 4th, it is as r — . The ratio 



a — ef L ' ' ' (l+ef 



of daily intensity of the northern, is to the southern then as one to ( ) ; or as 



1 to 1 — 4 e nearly; that is, 1 to 1 — j\. And the like ratio for the summer intensities 

 is as 1 to 1 + iV But J T is the extreme deviation for a few days only; the mean 

 between this and 0, or g 1 ^, would seem more correctly to apply to the whole seasons 

 of summer and winter. Taking then ^\ th of the greatest and least values of daily 

 intensity, Section IV, for the temperate zone, it appears that winter in the southern 

 hemisphere is now about 1° colder, and summer 3° hotter than in the northern hemi- 

 sphere. The intensities during spring and autumn may be regarded as equal in 

 both hemispheres. And the summer season of the south temperate zone being 

 hotter, is also shorter by about eight days, owing to the rapid motion of the earth 

 about the perihelion. 



In confirmation of these last deductions, the younger Herschel refers to the 

 glow and ardor of the sun's rays under a perfectly clear sky at noon, and observes, 



