As soon, however, as these eclipses of Jupiter’s moons began to be regularly observed, it was found that the intervals of time between two eclipses of any one moon were not equal, but exhibited a progressive variation in magnitude, and were longer by about 16 minutes and 26 seconds at one time of the year than at the other. The astronomer Roemer, in the year 1675, correctly concluded that this difference must be due to the fact that rays of light take time to traverse the earth’s orbit, and not to any want of regularity in the operation of this celestial timepiece. Hence, although the eclipses do happen at equal intervals of time, our information about them is delayed by the time taken for the ray of light to travel over the variable distance between Jupiter and our earth. These observations, critically considered, led, therefore, to the conclusion that the speed of light rays is about 186,500 miles a second. By means which it would take too long to describe here, experimental measurements of the velocity of light have been made many times since by various investigators by methods which do not involve astronomical observations, and the result has been to confirm the above value, and to give us a very exact knowledge of the speed with which rays of light travel through space. It is as shown in the table below:— THE VELOCITY OF LIGHT. Miles per second. From observations on Jupiter’s satellites (Roemer) 186,500 ” experimental measurements by Foucault (1862) 185,177 ” ” ” ” Cornu (1874) 185,487 ” ” ” ” ” (1878) 186,413 ” ” ” ” Michelson (1879) 186,364 ” ” ” ” ” (1882) 186,328 ” ” ” ” Newcomb (1882) 186,333