We have just spoken of the shadow as being elliptical; this was of course the case, the sun being quite low at the time, so that the round cone of darkness, technically known as the umbra, was cut very obliquely at the earth's surface. To realize the amount of this ellipticity or distortion, one would only need to hold some spherical body so as to cast a shadow on the ground about an hour and a half before sunset. The elongation was also continually increasing as the sun sunk toward the horizon, and its direction changed as the sun at the same time changed its direction or bearing, the longer axis of the ellipse always pointing toward the sun. This axis was, in Kentucky, about three hundred miles long; the shorter ninety; and this elliptical patch of darkness was moving in a course some thirty degrees south of east, or about twenty-three degrees south of its own longer diameter; its speed was about seventy-five miles a minute, or more than the average on the whole track, as before stated, and it required rather less than three minutes to pass any given point on the central line; this was consequently the duration of the totality; and short enough it certainly was, for the amount of work which was to be done by the observers. For the stations on or near the central line, it was important to obtain the absolute times of the contacts, and for this purpose transits were observed, to get the error and rate of the chronometer, for some time before and after the eclipse. The border observations locate the path on which the shadow travels, and determine its breadth; but to obtain the position of the shadow on this path at any fixed time, the true times of its arrival and departure at fixed points must be observed. But on the border no such preparations were necessary, only the interval being required; and a simple pendulum, without clock-work, was set up for this purpose, which broke the circuit at each second, and thus left its record, serving to count the number of seconds and the fraction between the beginning and end of the totality, which were observed and similarly recorded by means of a break-circuit key. This pendulum was so arranged as to break the circuit on the main telegraph line, and thus to be heard, and record its beats at a number of stations in different towns; but the main circuit did not itself mark upon the registers used by the observers, but mechanically (by means of what is called a relay magnet) broke short circuits set up at their stations, which could also be broken in another place by their own keys, without, of course, interfering with the main circuit itself; so that every observer could receive the pendulum beats upon his own record, without receiving those made by observers at other stations.