The stellar day, that is the time from one transit of a star to the next, is shorter than a solar day by 3_m._ 56_s._, so what is called sidereal time, regulated by the transits of well-known stars, in the manner we shall presently explain, by no means runs parallel with mean time so far as the clock indications go. Indeed when we look at a sidereal clock, we see something different to the clock we are generally accustomed to see. In the first place, we have twenty-four hours instead of twelve, and then generally there is one dial for hours, another for minutes, and another for seconds. That of course might happen in the case of the mean-time clock; but the mean-time clock is not often divided into twenty-four hours, although it formerly used to be, as the dials in Venice still testify. We now see the importance of an absolutely correct determination of the right ascension of stars; for this right ascension, expressed in hours, minutes, and seconds, is nothing more nor less than the time indicated by the sidereal clock, by the side of the transit instrument, when a star passes over, or transits, the central wire of that instrument. Hence it is the sidereal clock which keeps time with the stars, and which we keep correct by means of the transit instrument. [Illustration: FIG. 123.—System of wires in transit eyepiece. ] Let us show how this was always done some twenty or thirty years ago, and how it is sometimes done now. The transit room is kept so quiet that one can hear nothing but the ticking of the sidereal clock; the star to be observed is then carefully watched as it traverses the field of view over the wires, and the time of transit over each wire is estimated to the tenth of the time between each beat by the observer. We reproduce in Fig. 123 a rough representation of what is seen in the field of view of a transit instrument. Now if we could be perfectly sure of making an accurate observation by means of the central wire, it is not to be supposed that astronomers would ever have cared to use this complicated system of wires in their eyepieces; but so great is the difficulty of determining accurately the time at which a star passes a wire, that we have in eyepieces introduced a system of several wires, so that we may take the transit of the star first at one wire, then at another, until every wire has been passed over.