[ 350 ] 



LV. Remarks on the " Proposal fur estallishing a more correct 

 Account of Civil Time" published in our last Numler; with 

 a Suggestion for a still more correct Account. By Mr, 

 James Utting, of Lynn Regis. 



To Mr, Tilloch. 



Sir, — A OBSERVED in your valuable Journal for April 1820, 

 page 314, " A Proposal for establishing a more Correct Account 

 of Civil Time," affecting to be 720 less simple than the Gregorian 

 computation. 



That it is equally simple I acknowledge, but not that it is also 

 correct : the average tropical year to which your correspondent 

 W. W. approximates exceeds by more than eight minutes that 

 of the true solar year as used in our best solar tables. This error 

 in 500 years would amount to nearly three days, an error twenty 

 times greater than exists iu that which he attempts to correct. 



The following appears to come nearer to the point of accuracy 

 (which I discovered ten or twelve years since in the act of con- 

 structing a comparative table of years of different denominations) : 

 i.e. 900 revolutions of the sun* are performed in 32871S days 

 precisely, which is the least whole number contained in any pe- 

 riod of solar years constituting complete centuries. This period 

 contains seven days less than the Julian account : consequently, 

 by suppressing seven days in 900 years, the calendar would stand 

 corrected, i. e. every fourth and fifth centenary year alternately 

 would be leap years, but the intermediate centenary years would 

 constitute common years ; so that instead of suppressing three 

 days in four centuries as in tbe Gregorian account, we ought to 

 suppress seven days every 900 years: thus a correct calendar 

 may be obtained in reference to the solar year, admitting it in- 

 variable. But according to the celebrated M, Laplace, the tro- 

 pical year is found to diminish one second in about 460 years; 

 and admitting this decrease uniformly to continue, forty millions 

 of years would elapse before the error would accumulate to a 

 single day ! And as the accelerated motion of the earth in its 

 orbit has its limits, it would probably change to a retardation be- 

 fore the above per'od terminates, and a consequent increase in 

 the length of the solar year would obtain. Hence the above 

 correction of the calendar would undoubtedly be the most simple 



* The true solar or tropical year at the commencement of the present 

 century, as deduced from recent observations, contains iJ65 days five hours 

 forty-eight minutes forty-eight seconds. — Vide Laplace's Systeme du Monde, 

 and Vince's Astronomy, 3 vols, quarto, 2d edition (1814). 



and 



