TEMPERATURE AND TIME—MACGREGOR. 21 
other bodies are compared with the contemporaneous rotation of 
the earth relative to the fixed stars, the laws of their motions 
take forms which are simpler and more permanent than if any 
other motion be taken as standard. Hence, by common consent, 
the motion of the earth about its axis is taken as a standard 
with which other motions are compared. 
It is obvious that if the interval of time in which the earth 
makes a complete rotation were always the same, the laws of the 
motions of bodies, expressed by'reference to the contemporaneous 
rotation of the earth, would be identical with the laws of their 
motions, expressed in terms of time. Usually in stating the laws 
of the motions of bodies it is assumed that the rotation of the 
earth is uniform, and these laws are expressed in terms of time. 
But though the terminology of time is employed the laws of 
their motions are always really expressed in terms of the 
standard motion. 
Recent discussion of astronomical observations * seems to 
shew that the laws of the motions of heavenly bodies would 
take simpler forms, and would be more permanent, if the standard 
motion were that of an ideal earth, rotating so that its rate of 
rotation would slowly gain on the rate of rotation of the actual 
earth. If the time of the rotation of this ideal earth be assumed 
to be uniform, the time of the earth’s rotation, i. e., the sidereal 
day, must be regarded as increasing ata slow rate; and when the 
sidereal day is said to be increasing, nothing more is meant 
than that, as time goes on, a greater and greater number of 
rotations of this ideal earth occur during one rotation of the 
actual earth. We have no means of knowing whether the time 
of the rotation of the ideal earth is more or less variable than 
that of the real earth. But as the laws of the motions of bodies 
generally are simpler and more permanent when expressed in 
terms of the rotation of the ideal earth than when expressed in 
terms of the rotation of the real earth, it is convenient to assume 
the time of the former uniform and that of the latter variable. 
While therefore it is possible to describe instants of time with 
any degree of precision, it is not possible to measure the interval 
*See Thomson & Tait’s ‘‘ Treatise on Natural Philosophy,” Part IT , p. 830. 
