Historical origin
The day was subdivided
sexagesimally, that is by , by of that, by of that, etc., to at least six places after the sexagesimal point by the
Babylonians after 300 BC, but they did not sexagesimally subdivide smaller units of time. They did not use the hour, but did use a double-hour, a time-degree lasting four of our minutes, and a barleycorn lasting 3⅓ of our seconds (the ''helek'' of the modern
Hebrew calendar). The Egyptians had subdivided daytime and nighttime into twelve hours each since at least 2000 BC, hence their hours varied seasonally. The
Hellenistic astronomers
Hipparchus (''c.'' 150 BC) and
Ptolemy (''c.'' AD 150) subdivided the day sexagesimally and also used a mean hour (day), but did not use distinctly named smaller units of time. Instead they used simple fractions of an hour.
Medieval astronomers first subdivided the hour sexagesimally in 1200
[Seconds pendulum ] into ''pars minuta prima'' (first small part, our modern
minute), ''pars minuta secunda'' (second small part, our modern second), ''pars minuta tertia'' (third small part) and so on. Although a ''third'' for of a second remains in some languages, for example
Polish (''tercja'') and
Arabic (ثالثة), the modern second is now subdivided decimally.
The second first became measurable with the development of
pendulum clocks keeping ''mean time'' (as opposed to the ''apparent time'' displayed by sundials), specifically in 1670 when William Clement added a
seconds pendulum to the original pendulum clock of
Christian Huygens.
[Long Case Clock: Pendulum ] The seconds pendulum has a period of two seconds, one second for a swing forward and one second for a swing back, enabling the
longcase clock incorporating it to tick seconds.
In 1956 the second was defined in terms of the period of revolution of the
Earth around the Sun for a particular
epoch, because by then it had become recognized that the Earth's rotation on its own axis was not sufficiently uniform as a standard of time. The Earth's motion was described in
Newcomb's Tables of the Sun, which provides a formula for the motion of the Sun at the epoch 1900 based on astronomical observations made between 1750 and 1892.
The second thus defined is
:''the fraction 1/31,556,925.9747 of the
tropical year for 1900 January 0 at 12 hours
ephemeris time.''
This definition was ratified by the Eleventh General Conference on Weights and Measures in 1960. The ''tropical year'' in the definition was not measured, but calculated from a formula describing a mean tropical year which decreased linearly over time, hence the curious reference to a specific ''instantaneous'' tropical year. Because this second was the independent variable of time used in
ephemerides of the Sun and Moon during most of the twentieth century (Newcomb's Tables of the Sun were used from 1900 through 1983, and
Brown's Tables of the Moon were used from 1920 through 1983), it was called the ''ephemeris second''.
With the development of the
atomic clock, it was decided to use atomic clocks as the basis of the definition of the second, rather than the revolution of the Earth around the Sun.
Following several years of work,
Louis Essen from the
National Physical Laboratory (Teddington, England) and
William Markowitz from the
United States Naval Observatory (USNO) determined the relationship between the hyperfine transition frequency of the
caesium atom and the ephemeris second.
Using a common-view measurement method based on the received signals from
radio station WWV,
they determined the orbital motion of the
Moon about the Earth, from which the apparent motion of the Sun could be inferred, in terms of time as measured by an atomic clock. As a result, in 1967 the Thirteenth
General Conference on Weights and Measures defined the second of
atomic time in the
International System of Units as
:''the duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom.''
During the 1970s it was realized that
gravitational time dilation caused the second produced by each atomic clock to differ depending on its
altitude. A uniform second was produced by correcting the output of each atomic clock to
mean sea level (the rotating
geoid), lengthening the second by about 1. This correction was applied at the beginning of 1977 and formalized in 1980. In relativistic terms, the SI second is defined as the
proper time on the rotating geoid.
[R. A. Nelson ''et al.'', "", ''Metrologia'' 38 (2000) 509-529, p. 515.]
The definition of the second was later refined at the 1997 meeting of the
BIPM to include the statement
:''This definition refers to a caesium atom at rest at a temperature of 0'' K.
The revised definition would seem to imply that the ideal atomic clock would contain a single caesium atom at rest emitting a single frequency. In practice, however, the definition means that high-precision realizations of the second should compensate for the effects of the ambient temperature (
black-body radiation) within which atomic clocks operate and extrapolate accordingly to the value of the second as defined above.