time. Symbol d. Any of the various periods associated with one rotation of Earth. As explained under time, there are various lengths of day. The obvious day is the time between successive high noons or other sundial readings, termed the apparent solar day. Because of the parameters of Earth's elliptical orbit, the length of such period varies up to 30 seconds from the nominal 24 hours, the greatest discrepancy being reached as Earth approaches perihelion (its closest point to the Sun, currently near the beginning of January, but nearly half an hour earlier each orbital period; see year). The averaged value, the mean solar day, has long been defined as 24 hours.
The routine offset value of a sundial from a regular clock for any particular day is computed by the equation of time. The cumulative effect currently brings the sundial a maximal 16 minutes 23 seconds fast in early November and a maximal 14 minutes 20 seconds slow in mid-February. Around those dates and in mid-May and late July the length of the apparent solar day is the average 24 hours. Sundial readings are correct in mid-April and mid-June, near the start of September and in late December. In June, and more so in December, the solar days are longer than the mean as the solstice is passed, causing the apparent paradox of sunrise being later as the daylight lengthens for the winter hemisphere, and of sunset being later as the daylight shortens for the summer hemisphere (an anomaly that can persist for a week), and is mirrored in a like period preceding the solstice.
Because of the friction of tidal effects (applicable to land as well as water), the rotational speed of Earth is progressively slowing, lengthening the mean day by nearly 15μs per year, using the second now fixed by the atomic clock. The size of that second was set to the second of ephemeris time, related to the year 1900 but equated with 1/86400 of the mean solar day as observed over the period 1750-1892. Thus it is'effectively 1/86400 of the mean solar day around the year 1820, making the like day of 2001, by accumulation of all those microseconds, about 86 400.0028 s long. Over 365 days, the excess amounts to about 0.75 s, which prompts the inclusion of leap seconds in Universal Time.
Since Earth is simultaneously travelling about the Sun, with the same anti-clockwise turning viewed from the (North) Pole Star as its rotation, it has to turn fractionally more than one rotation to bring the Sun back to a corresponding position each day, by the amount that offsets over a year the effective one counter-rotation of orbiting the Sun. Hence the sidereal day, the time for one revolution relative to the stars and of which there are 366¼ in a year, is less than the apparent solar day by about 1/366 day, i.e. almost 4 minutes. The variation of the sidereal day itself is a very minor fraction of a second, so the prefix ‘mean’ is usually omitted. Current values are as follows:
[The Astronomical Almanac for the Year 2002 (Washington: US Government Printing Office and London: HMSO, 2001), also Explanatory Supplement to the Astronomical Almanac (Mill Valley, CA: University Science Books, 1994).]
| 1 sidereal day | = 0.997 269 566 33~ mean solar day |
| 1 mean solar day | = 1.002 737 909 35~ sidereal day |
| 1 sidereal day | = 23.934 470~ mean solar hours |
| = 23 hours 56 minutes 4.090 53~ seconds (mean solar) | |
| 1 mean solar day | = 24.065 710~ sidereal hours |
| = 24 hours 3 minutes 56.555 37~ seconds (sidereal) |
| second | |||||
| 60* | minute | ||||
| 3 600 | 60 | hour | |||
| 86 400′ | 1440 | 24 | day | ||
| 7 | week | ||||
| 28 | 4 | vernacular ‘lunar month’ | |||
| 30± | solar calendar month | ||||
| 365± | 52± | 12 solar calendar year | |||
| The ± indicates approximate values; * see second for exceptions. |
In European tradition, the day starts at midnight, i.e. such that high noon is its midpoint. However, this demarcation is not universal, even within European practice. Astronomers, for obvious reasons, favoured noon as the start (see astronomical day system as well as calendar for the overall sequential identification of days), and such was also long the practice at sea, the former a half-day behind, the latter the same ahead. (The dating of Capt. James Cook's initial activities in Australasia presents an interesting example. His log changed date at noon, normally putting it ahead of the familiar date. However, having travelled westward from Britain without advancement for any dateline, he was, by then, a day behind overall. Thus his afternoons corresponded to the familiar date, his mornings were behind.)
By natural extension, the term day is used, appropriately qualified for the equivalent periods, for other planets, e.g. the Martian day.
See also degree-day.




