212 M. Arago on the Thermometrical State 



on the subject, it may be satisfactory to examine how this sidereal 

 day is determined. 



In every observatory there is a wall, carefully built and fi- 

 nished, or something equivalent, extending with the greatest 

 accuracy from south to north. The astronomer who wishes 

 to see if his pendulum is regulated by sidereal time, notes, with 

 all possible precision, the moment when a star is seen along the 

 length of the wall. The next day he repeats the observation 

 upon the same star. If twenty-four hours, neither more nor 

 less, have elapsed between the first and second observation, the 

 pendulum is regulated. It is going too fast or too slow, when, 

 between the two transits of the star in the line of the wall, the 

 hands have marked more or less than twenty-four hours *. 



The ancients must have regarded the sidereal day as the mea- 

 sure, in time, of the rotation of the celestial sphere, because 

 they believed that the earth was immoveable. The moderns 

 have demonstrated that the earth revolves ; and hence, that 

 when the star appears to place itself in the line of the meridian 

 wall, it is in reality the wall which comes to meet the star. They 

 were therefore inevitably led to see, in the sidereal day, the du- 

 ration of the rotation of our globe. 



We have thus brought the question of temperature, which we 

 wish to solve, to a pioblem of the measure of time, because the 

 ancients were not acquainted with the thermometer. Will any 

 one ask, what has been gained by this substitution, seeing that 

 the ancients no more possessed pendulums ; or, at all events, 

 none of them have come down to our times ? True : but we 



• Respecting a pendulum regulated according to sidereal time, it ought to 

 be noted, that such a pendulum, indicating that exactly twenty-four hours 

 have elapsed between the two consecutive transits of a star by the meridian 

 wall, would indicate 24'> 3' 56'' during the solar day of mean time in common 

 use. 13 ut this difference may be easily explained. Let it be supposed that 

 the sun and a certain star are in the same region of the heaven ; that is to 

 say, let it be admitted, that, as on this day, the two luminaries pass at the 

 same moment the line of the meridian wall. To-morrow, on the return of 

 the star ; in other words, when the sidereal day shall have been finished, the 

 sun will no longer be in the same position as with the star, it will be more to 

 the east ; it wiU not arrive at the line of the wall, till after all the points of 

 the arc by which it is removed, shall have passed the meridian. Now, the 

 time required for the arc of the sun, so displaced, during a day, wholly to tra- 

 verse the meridian; is according to its mean value, 3' and 56". 



