SUN ? S TEANSITS. 



59' S'2". The sun, therefore, will not yet have come to the 

 meridian, and will not arrive at it until it is carried by the 

 diurnal motion of the firmament through this space Fig. 3. 

 of 59' 8 "2". But since the firmament moves at N 



the rate of 15' in each minute of time, it will take 

 3 m 56 s to carry the sun, s' to the meridian. It 

 follows, therefore, that, supposing the sun to move 

 daily 59' 8-2" eastward at right angles to the 

 meridian, a solar day would exceed a sidereal day 

 by 3-" 56". E 



If the eastward daily motion of the sun, measured 

 at right angles to the meridian, were uniform there- 

 fore, the interval between its successive transits 

 would possess all the requisites for a chronometric 

 unit, and although the solar day would not be 

 equal in length to the sidereal day, it would, nevertheless, be of 

 invariable length, and would besides be in complete accordance 

 with those periodical vicissitudes of light and darkness which 

 have been, by common consent, used by mankind in all countries 

 and in all ages as the measures of time. 



But the solar day is wanting in fact in this essential condition ; 

 it is not invariable in length. Its variation, though not great, is 

 nevertheless such as to render it unsuitable as an unit of time, 

 even for civil, to say nothing of astronomical, uses. No clock or 

 watch could be constructed which would continue to go with the 

 sun. A clock, which at one time of year would correspond with 

 the meridional transits, would at another either anticipate them, 

 or fall behind them. 



The variation in the rate at which the sun is displaced daily 

 towards the east, and at right angles to the meridian, arises from 

 several causes. First, the rate at which the sun moves upon the 

 firmament is subject to variation. While its average daily dis- 

 placement is, as we have stated, 59' 8*2", it amounts at the 

 beginning of the year to 1 V 9'9", and at the middle of the year 

 to only 5V 11-5". Although we are not directly concerned here 

 with the cause of this variation, it may be as well to observe, that 

 it arises from the fact that the earth does not revolve in an exact 

 circle with the sun in the centre, which it must have done if its 

 motion were uniform, but in an oval, the sun being nearer to one 

 end than to the other, and the rate of the motion increasing as the 

 distance of the sun decreases. Secondly, the motion of the sun is 

 not generally at right angles to the meridian, but more or less 

 oblique to it at different seasons, and the more oblique it is 

 to the meridian, the less does a given displacement affect its 

 eastward motion at right angles to the meridian. Thirdly, the 



K2 131 



