History of Leap second

	Revision 7 . . November 27, 2001 10:49 am by (logged).111.13.xxx
	Revision 6 . . November 24, 2001 2:00 am by AxelBoldt [Explicit URLs (to make printed out version of article more useful)]
	Revision 5 . . (edit) November 24, 2001 12:52 am by Tompeters [some clarifications and more links]
	Revision 4 . . November 23, 2001 8:29 pm by ManningBartlett [copyediting an already excellent article]
	Revision 3 . . November 23, 2001 8:07 pm by Tompeters [Completely rewritten because the original author obviously misunderstood the pertinent issues.]
	Revision 2 . . (edit) November 23, 2001 5:24 pm by Juuitchan
	Revision 1 . . November 23, 2001 5:24 pm by Juuitchan

---

Difference (from prior major revision) (no other diffs)

Changed: 5c5

The instruction to insert a leap second will be given whenever the difference between UTC and GMT becomes 0.9 s. The insertion only ever occurs at the end of 30 June or 31 December. It is implemented as follows: after clock time 23:59:59, an additional second at 23:59:60 is counted, before the clock jumps to 0:00:00 of the next day.

The instruction to insert a leap second will be given whenever the difference between UTC and GMT becomes 0.9 s. The insertion only ever occurs at the end of 30 June or 31 December. (According to the official rules, the IERS can also insert leap seconds at the end of the first or third quarter, although the IERS has never done so in practice.) It is implemented as follows: after clock time 23:59:59, an additional second at 23:59:60 is counted, before the clock jumps to 0:00:00 of the next day. (The IERS can also omit a second at the above mentioned points instead, if the difference between UTC and GMT is sufficently negative -- however, this has never happened.)