The arrival of the new year will take a bit longer this winter. The reason: bringing the calendar back into synchronization with the earth's rotation requires the addition of an extra second. Leap seconds are not new: 22 have been added since 1972, on June 30 or December 31. But under proposals to be discussed at a November meeting of the radiocommunication sector of the International Telecommunication Union--the United Nations agency in charge of broadcasting official time--this leap second might be the last. Astronomers, historians and others are upset about the prospect of losing the leap second and have initiated a spirited debate to keep time tied to the earth's rotation.

How long can people spend arguing about a second? "It's been going on for about five years," says Judah Levine, a physicist in the time and frequency division at the National Institute of Standards and Technology (NIST) in Boulder, Colo. Word of the proposed abolition became public in July, thanks to an e-mail tip to timekeeping officials by Daniel Gambis, head of the Earth Orientation Center of the Frankfurt, Germany-based International Earth Rotation and Reference Systems Service (IERS), which decides when to add leap seconds.

Leap seconds are needed because the earth's spin is slowing down, gradually and unevenly. The rotational changes arise because of tidal forces exerted by the moon and inertial effects related to the liquid outer core sloshing around and to the cycle of evaporation, in which water at the equator gets deposited at the poles as ice that melts seasonally.

The present system is a compromise between taking advantage of the most accurate timepieces--that is, atomic clocks--and respecting traditional timekeeping via the sun's position. To people who want to end leap seconds, Levine explains, "the really fundamental quantity is not time but frequency--and frequency comes from quantum mechanics; it is a property of atoms. And what these folks really want is for time to represent frequency in a smooth, continuous way." Levine does not speak officially for NIST, but he is the person who, on December 1, will formally alert authorities to add the leap second at the end of the month.

The primary advocates of a frequency standard are people behind modern telecommunications systems. GPS, for example, does not adjust for leap seconds--a problem made apparent in 2003, when the length of time since the last leap second caused the clocks in some GPS receivers to malfunction so that they gave the time as 62:28:15.

To illustrate the issue posed by leap seconds, Levine points to navigation. "Every time there's a leap second, the thing that's moving continues to move, but the clock stops," he explains. "So the people who deal with physical processes do not want leap seconds." Tom O'Brian, chief of NIST's time and frequency division, adds that leap seconds occur unpredictably. The IERS did not decide on this year's leap second, the first in seven years, until July 2005. To keep clocks from drifting too far from the day-night cycle, abolitionists would presumably need to add, say, several leap minutes every few hundred years.

The existing compromise system, Levine notes, also sows confusion. For one, the leap second occurs in the middle of the day in Asia and Australia, causing a time hiccup during stock trading. For another, the more timescales there are, the easier it is for a programmer to make an error in calculations.

Astronomers are deeply dismayed at the prospect, which would decouple time from the earth's rotation, much as going off the gold standard did for currency. "We need to know where the earth is pointing so we can point a radio" for spacecraft communications, says Mike Hapgood, the geophysical secretary of Britain's Royal Astronomical Society. Modifying equipment, he argues, could become expensive.

Not every astronomer thinks that adjustments will be arduous. Cornell University's Donald Campbell, former director of the center that runs the Arecibo radiotelescope, states that rewriting the software to adjust for the difference between atomic time and universal time is not a big deal.