must close the cap at eighty-nine; which leaves him eleven seconds in which to get the last exposure of two seconds. If, however, there has been an unfortunate delay anywhere, and the time-caller is calling eighty-five instead of seventy-nine, the astronomer realizes that if he gives the full ten seconds up to ninety-five, it will be practically impossible for him to get the last exposure of two seconds unless the eclipse lasts unexpectedly long (as may happen); he can thus shorten the exposure from ten seconds to five seconds if he likes, so as to make sure of the last plate. He has probably thought over what to do in such a case beforehand, for it is desirable not to leave such decisions to be made in the exciting moments of the eclipse.
Now I think I have explained the details, and the time of total eclipse is drawing very near, as you see. The daylight is perceptibly less (a few lamps were turned out), but it is by no means quite dark at a total eclipse; one can read the figures on a watch-face, though the time-caller may prefer to use a metronome which gives him the seconds by ear, and leaves his eyes free to watch the eclipse. By this time we ought to be feeling distinctly chilly; it has been said that a cold wind springs up about now, though others say the feeling is merely due to the fall of temperature. You know well enough the way in which it seems cooler even when the Sun only goes behind a cloud; at a total eclipse this effect is much more noticeable; and it must be admitted that there is a very solemn feeling about it all. Natives are often very much frightened; if they have not been told anything beforehand it