How We Calculate

The Japanese zodiac (eto) uses the same twelve animals as the Chinese one, but since 1873 Japan has counted them by the Western calendar. That year the Meiji government replaced the old lunisolar calendar with the Gregorian one, and the animal of the year came to follow the civil year: it changes on 1 January, not at the lunar New Year. Our calculation follows that convention, and we state it as such: the animal is the one of the Gregorian year of your birth, counted in the usual twelve-year cycle in which 2000 was a Dragon year.

Because every date in the calculator falls after 1873, no birth is affected by the older lunisolar reckoning. The rule is simple enough to be exact by construction, and our tests check it at the turn of the year, where a mistake would show: 31 December 1999 is a Rabbit and 1 January 2000 a Dragon. The cycle is fixed, so no other year can drift: once the turn of the year is right, every year is.

This is also why a Japanese and a Chinese person born on the same day in January or early February can have different animals: the Chinese animal changes at the lunar New Year, the Japanese one on 1 January.

BaZi, the "eight characters", describes a birth by four pillars: year, month, day and hour, each written with one of the sixty pairs of a heavenly stem and an earthly branch. The calculator shows the day pillar, the one that practitioners read as the core of the chart, because it depends only on the date and can be computed without approximation. The four-pillar method itself is set out in classical works such as Wan Minying's Sanming tonghui of the Ming dynasty.

The sixty-day cycle has run without interruption for more than two thousand years. China's national standard for the calendar, GB/T 33661-2017, fixes its reference point: the day that corresponds to 1 October 1949, Beijing time, is a jiazi day, the first of the cycle. From that day we count forward or backward in steps of sixty. Our tests check this anchor and compare every day from 1900 to 2100 with an independent implementation of the Chinese calendar; they agree on all of them.

One convention needs stating. The Chinese double hour zi runs from 11 p.m. to 1 a.m., across midnight, so a birth between 11 p.m. and midnight falls in a zi hour shared by two civil dates. We follow the national standard and count the day from civil midnight. If you were born in that hour, a practitioner who starts the day at 11 p.m. would read the following day's pillar.

The tzolk'in is the Maya count of 260 days: thirteen numbers combined with twenty named days, so that each day has a name such as 4 Ajaw and the same combination returns every 260 days. It ran continuously alongside the 365-day haab, and together they dated events on Maya monuments.

Turning a Maya date into a European one needs a correlation, a fixed number that links the Maya count of days to the Julian day count. We use the most widely accepted value, the GMT correlation 584283, under which the end of the thirteenth b'ak'tun, 13.0.0.0.0 4 Ajaw 3 K'ank'in, fell on 21 December 2012. In 2013 Kennett and colleagues tested the correlation by radiocarbon dating a carved wooden lintel from Tikal with a known Maya date, and their results supported the GMT family. Martin and Skidmore had argued in 2012 for 584286, three days later; we state the choice because a different correlation would move every day sign.

We also start the Maya day at midnight, a convention: the Maya sources do not tell us when their day began. The calculation is otherwise exact. It reproduces the sample dates of Reingold and Dershowitz's Calendrical Calculations, and our tests check every day from 1900 to 2100 against an independent implementation, with 21 December 2012 as 4 Ajaw.

The haab is the Maya year of 365 days: eighteen months of twenty days, followed by five days called Wayeb that stood outside the months. It has no leap year, so it slowly drifts against the seasons, and Maya scribes paired it with the 260-day tzolk'in to give each day a fuller name.

We use the same correlation as for the tzolk'in, the GMT value 584283, which places the end of the thirteenth b'ak'tun on 21 December 2012, the day 4 Ajaw 3 K'ank'in. The radiocarbon study by Kennett and colleagues in 2013 supported this family of correlations; Martin and Skidmore's proposal of 584286 would move every haab date by three days, which is why we state the value we use.

Your haab sign is the month in which your birthday falls, or Wayeb for the five closing days. As with the tzolk'in, the day is counted from midnight by convention. The calculation is exact once the correlation is fixed: it reproduces the sample dates published in Reingold and Dershowitz's Calendrical Calculations, and our tests check 21 December 2012 as the month K'ank'in and every day from 1900 to 2100 against an independent implementation.

Because the haab has no leap day, the same month falls a quarter of a day earlier against our calendar every year: two people born on the same date four years apart can have neighbouring haab days.

The Aztec tonalpohualli is a count of 260 days, like the Maya tzolk'in: thirteen numbers combined with twenty day signs, such as Coatl (serpent) or Ehecatl (wind). Your sign is the day sign of your birth.

To place it on our calendar we need one firm link between the Aztec count and a European date. The best attested is the fall of Tenochtitlan, 13 August 1521 in the Julian calendar, recorded as the day 1 Coatl. This is the correlation established by Alfonso Caso and set out in the calendrical appendix of the UNAM edition of the Crónica mexicáyotl edited by Gabriel Kruell. It is consistent with the Maya GMT correlation, so the Aztec and Maya day counts stay aligned in our calculator.

As for the Maya, we count the day from midnight; this is our convention. Our tests check the correlation directly, with 1 January 2000 as the day sign Ehecatl, and compare every day from 1900 to 2100 with an independent implementation.

The colonial sources record that different towns did not always keep the count in step, so a single correlation is a choice. We use the one tied to the best documented date, and we state it here so that anyone who prefers another alignment knows exactly what we did.

The Balinese pawukon is a cycle of 210 days divided into thirty weeks of seven days, each week called a wuku and each with its own name, protective deity and character in Balinese tradition. The cycle has no year: it simply repeats, and the great festival of Galungan returns every 210 days, on the Wednesday of the week Dungulan. We show your wuku as a fact about your birth, the week of the pawukon in which you were born.

The arithmetic of the cycle is set out in Reingold and Dershowitz's Calendrical Calculations, which gives the order of the thirty wuku and a formula for any date. We also checked it against an official date: the government of Buleleng announced its holidays for Galungan on 16 to 18 June 2026, and our calculation places 17 June 2026 in Dungulan, as it must. Our tests also check that the wuku stays the same through each seven-day week for every day from 1900 to 2100.

The week itself needs no convention: once the day is known, the wuku follows from the 210-day count without approximation, and no time or place of birth is needed. The calculation also reproduces all the Balinese sample dates printed in Calendrical Calculations.

The Medicine Wheel astrology on this site is the system published by Sun Bear and Wabun in 1980 in The Medicine Wheel: Earth Astrology. It divides the year into twelve moons, each with a totem animal, a plant, a mineral and a colour, and the moon of your birth gives your totem.

We take the moons and their dates exactly as the book prints them. Each moon runs between two fixed calendar dates, and both ends are inclusive, as the book gives them. 29 February belongs to the Cougar, the moon of late winter in which it falls. Our tests check every boundary printed in the book, so that a birth on the first or last day of a moon gets the totem the authors assigned.

It is important to say what this system is. The Medicine Wheel as a calendar of birth totems is the authors' own work, written for a general readership in the twentieth century; it draws on Native American imagery but is not the calendar of any particular nation, and the book does not claim otherwise. We present it as Sun Bear and Wabun's system, with their dates, and we do not add to it.

Because the moons are tied to fixed dates rather than to the sky, no time or place is needed and the calculation is exact for every birth date.

In Thailand the animal of the year does not change on 1 January or at the Chinese New Year, but on the first day of the waxing moon of the first lunar month of the Thai royal calendar, which falls between November and December. The Department of Provincial Administration, which keeps the civil register, states this rule, and we follow it on the civil date of birth, as the register does.

To find that day we need the Thai lunar calendar, whose years are of three kinds: ordinary years of 354 days, years with an extra day, and years with a second eighth month. Which kind each year is, is set by the calendar promulgated for official use, and different calendars do not always agree. For 1859 to 2027 we use the official sequence: the table published by Loy Chunpongtong in the Journal of the Royal Institute of Thailand, completed by the official dates announced up to 2027. Our tests reproduce all 69 official dates we collected, such as the Buddhist holidays announced by the Bank of Thailand, and the register's own example: a child born on 28 November 2019, the second day of the first month, is a Rat.

From 2028 there is no official table yet, so the calculator predicts the kind of each year with the rule described by J. C. Eade in The Calendrical Systems of Mainland South-East Asia. This is a declared approximation: over the official years the rule places the extra day differently in about one year in four, and the second eighth month differently twice in 169 years. After 2027 the day the animal changes could therefore be one day off, and in those rare years a whole month off.

The Celtic tree calendar has thirteen months of 28 days, each named after a tree, from Birch to Elder, beginning on 24 December. Your sign is the tree of the month of your birth.

It is essential to say where this calendar comes from. It was set out by the poet Robert Graves in The White Goddess, first published in 1948; we use the dates of the third edition of 1952. Graves built it from medieval Irish tree-alphabet lore and presents it as his own reconstruction, and there is no ancient Celtic source for a calendar of thirteen tree months with these dates. We present it as Graves's calendar.

We follow Graves exactly, including his one oddity: the thirteen months of 28 days cover 364 days, and the remaining day, 23 December, belongs to no tree. A birth on 23 December therefore gets no Celtic card; we do not invent one. 29 February falls inside the month of the Ash, as the dates require.

Our tests check every boundary of the thirteen months, both ends included, and the special day of 23 December. Because the months are tied to fixed dates, no time or place is needed and the calculation is exact.

The Chinese year animal changes at the lunar New Year, which falls between 21 January and 20 February. The year begins with the second new moon after the winter solstice, or the third when a leap month intervenes, with every new moon and solar term reckoned in Beijing time. These are the rules of China's national standard for the calendar, GB/T 33661-2017, and of the algorithm in Reingold and Dershowitz's Calendrical Calculations, which also uses Beijing local mean time before 1929.

We compute the New Year from the positions of the Sun and Moon rather than copying a table, and then check it against the official tables. The Hong Kong Observatory publishes conversion tables for every year from 1901 to 2100: our calculation matches all 200 New Year dates. It also matches the New Year dates given in Calendrical Calculations for the first years after 2100.

Your animal is the one of the lunar year that contains your date of birth. We compare the New Year date with your local civil date of birth, as is usual in practice; this is our convention, and it matters only to someone born on the New Year's day itself far from China. Someone born on 9 February 2024 is a Rabbit, someone born on 10 February 2024 a Dragon.

Because the New Year depends on astronomy, this sign appears a moment after the others: the calculator loads its astronomy after you enter the date.

The Korean year animal follows the Korean lunisolar calendar, which uses the same rules as the Chinese one: the year begins with the second new moon after the winter solstice, or the third in years with a leap month. The difference is the meridian. Korea reckons its calendar in Korean time, which has been 135° east for most of the last century, but 127°30′ east in 1908 to 1911 and 1954 to 1961, and local mean time before 1908. Because a new moon can fall just before midnight in Beijing and just after midnight in Seoul, the Korean New Year sometimes falls a day, or a month, away from the Chinese one.

The Korea Astronomy and Space Science Institute publishes the official Korean calendar. Our calculation matches all 151 of its New Year dates from 1900 to 2050, and it finds exactly the nine years between 1901 and 2050 in which the Korean New Year differs from the Chinese, as listed by Park, Mihn and Ahn in their study of the Korean calendar. In 2027, for example, the Korean New Year falls on 7 February and the Chinese on 6 February, so a child born on 6 February 2027 in Korea is still a Horse.

As for the Chinese sign, we compare the New Year with your local civil date of birth. After 2050 there is no official table yet; our calculation then agrees with an independent computation, which we use as a check, not as an official source.

The Vietnamese year animals are the Chinese twelve with two differences in name: the Cat takes the place of the Rabbit and the Buffalo that of the Ox. The year begins at Tết, the lunar New Year, computed with the same rules as the Chinese calendar but at Vietnam's own meridian, as described in Reingold and Dershowitz's Calendrical Calculations.

The meridian matters. North Vietnam reckoned its calendar in UTC+8 until 1967 and in UTC+7 from 1968, after a government decision of 1967. A new moon near midnight can therefore fall on different days in Hanoi and in Beijing, and Tết can come a day or even a month away from the Chinese New Year. In 1985 Tết fell on 21 January, a month before the Chinese New Year of 20 February, so a child born in Vietnam between those dates was already a Buffalo while a child born in China was still a Rat.

We calculate for Hanoi, and we state one simplification. South Vietnam kept UTC+8 until 1975, so in two years its Tết differed from the North's: for births in the South on 29 January 1968 or 16 February 1969, our animal is the North's, one year later than the one used in the South at the time. Our tests check Tết 1968 on 29 January and Tết 1985 on 21 January.

The Western sun sign is the sign of the zodiac in which the Sun stood at the moment of your birth. The zodiac used is the tropical one, whose twelve signs of 30° are counted from the March equinox: Aries begins where the Sun crosses the celestial equator in spring. This definition goes back to Ptolemy's Tetrabiblos, which counts the signs from the solstices and equinoxes, and it is the one used by Western astrology today.

We compute the Sun's apparent longitude for the exact instant of your birth, with the astronomical methods described in Jean Meeus's Astronomical Algorithms, and take its sign. There are no fixed date ranges: the Sun enters each sign at a different moment each year, which is why horoscope columns disagree about the boundary days. Our tests check that the change from Pisces to Aries in 2000 falls at the March equinox published by the US Naval Observatory, within two minutes.

On most days everyone born that day has the same sign, whatever the time and place, and we show it from the date alone. On the two or three days a month when the Sun changes sign somewhere in the world during the day, we ask for the time and place of birth before showing it, rather than guessing.

The sidereal zodiac is anchored to the stars rather than to the equinoxes. Because the Earth's axis slowly precesses, the March equinox drifts backwards against the constellations by about one degree every 72 years; today the tropical signs stand nearly a whole sign away from the star groups that gave them their names. Sidereal astrology corrects for this drift by subtracting an angle, the ayanamsa, from the Sun's tropical position.

We use the Fagan–Bradley ayanamsa, defined by Cyril Fagan and Donald Bradley and set out in Fagan and Firebrace's Primer of Sidereal Astrology: its value at 1950 is carried forward and backward with the precession of the equinoxes, and we use the Sun's true position, including nutation. The result can differ from the tropical sign by one sign: someone born on 1 May 1990 is a tropical Taurus but a sidereal Aries.

Our tests check that the sidereal sign changes exactly when the Sun's tropical longitude reaches the ayanamsa. As for the Western sign, we show it from the date alone on the days when no time of birth could change it, and ask for the time on the others.

Other sidereal traditions use different ayanamsas; the Vedic sign on this site, for example, uses the Indian government's fixed value, while the Fagan–Bradley ayanamsa moves with precession: the gap between them grows from a tenth of a degree in 1900 to almost two degrees today, so the two signs can disagree for births near a boundary.

Vedic astrology, Jyotish, uses a sidereal zodiac, fixed to the stars. The sign we show here is the solar sign, the rashi in which the Sun stood at birth, with the Sanskrit names Mesha to Meena.

To pass from the tropical to the sidereal zodiac an Indian astrologer subtracts the ayanamsa, and the choice of its value has a history. In 1955 the Calendar Reform Committee appointed by the Government of India studied the many values in use and recommended a single standard for the national calendar. We use the committee's recommended fixed value of 23° 15′ for the solar sign, so that the sign follows the official Indian standard.

We compute the Sun's tropical position for the instant of birth, subtract 23° 15′, and take the sign of 30° in which the result falls. Our tests check that the sign changes from Meena to Mesha exactly when the Sun's tropical longitude reaches 23° 15′.

As for the Western sign, we show the Vedic sign from the date alone whenever no time of birth could change it, and ask for the time on the days near a boundary. The Moon's nakshatra, which Vedic astrology also reads at birth, has its own section below and always needs the time.

The Persian signs are the twelve signs of the zodiac under their Persian and Arabic names, from Hamal (the Ram) to Hut (the Fish), as they appear in the astronomy and astrology of the Islamic world. Al-Bīrūnī's Book of Instruction in the Elements of the Art of Astrology, written in 1029, lists them and describes the zodiac as divided into twelve equal signs counted from the spring equinox.

We therefore compute this sign exactly as the Western one: the sign of 30° in which the Sun's apparent longitude stood at the moment of birth, counted from the March equinox. That the medieval Persian tradition used the tropical zodiac is our reading of al-Bīrūnī, and we state it as such. In practice your Persian sign always corresponds to your Western sign under its Persian name: someone born on 1 May is a Western Taurus and a Persian Saur, the Bull.

Like the Western sign, it is shown from the date alone on the days when no time of birth could change it, and the calculator asks for the time on the two or three days a month when the Sun changes sign. Our tests check the sign changes at the moment the Sun's longitude crosses a multiple of 30°.

The zodiac of twelve equal signs was invented in Babylon, in the second half of the first millennium BC, and the earliest horoscopes are Babylonian. Unlike the later Greek zodiac it was tied to the stars: its signs were measured from fixed reference stars, so it is sidereal. Francesca Rochberg's edition of the Babylonian horoscopes shows how births were recorded, and John Britton's study of the introduction of the uniform zodiac reconstructs where its signs began.

We follow Britton's reconstruction. He derived the offset between the Babylonian zodiac and the equinox for the period when the zodiac was introduced, with an uncertainty of about a tenth of a degree, and we carry that offset forward to the date of your birth with the precession of the equinoxes. Today the Babylonian signs stand about 25° away from the tropical ones: on 1 May 2026 the Sun is about 11° into tropical Taurus but in the Babylonian sign of the Hired Man, our Aries.

The tenth of a degree of uncertainty matters only for births within a few hours of a sign change, and we state it. As for the other solar signs, the Babylonian sign is shown from the date alone when no time could change it, and the calculator asks for the time on the days near a boundary.

The nakshatras are the 27 lunar mansions of Indian astronomy: the circle of the zodiac divided into 27 equal parts of 13° 20′, through which the Moon passes in about a month. The nakshatra in which the Moon stood at birth is one of the most important elements of a Vedic chart, as Varāhamihira's Bṛhat Saṃhitā already describes in the sixth century.

Because the Moon moves about 13° a day, it changes nakshatra roughly once a day, so this sign always needs the time of birth. We compute the Moon's apparent position for the instant of birth and subtract the ayanamsa recommended in 1955 by the Calendar Reform Committee of the Government of India: 23° 15′ at 21 March 1956, carried forward and backward with the precession of the equinoxes and including nutation, so that the mansions stay fixed to the stars.

Our tests check sample births, for example the Moon in Shravana on 15 March 1950 at 06:00 UT, and they check that the boundary between two mansions falls exactly where the Moon's sidereal longitude reaches a multiple of 13° 20′.

Some Indian traditions use other ayanamsas, which can move a birth near a boundary into the neighbouring nakshatra; we use the committee's value because it is the official standard.

The ascendant, or rising sign, is the sign of the zodiac that was rising on the eastern horizon at the moment and place of your birth. It changes about every two hours, faster or slower depending on the sign and the latitude, so it needs both the exact time and the place.

In geometrical terms the ascendant is the point where the ecliptic, the Sun's apparent path, crosses the eastern horizon. We compute it from the local sidereal time and the latitude of the birthplace with the standard formula of spherical astronomy, as given in Meeus's Astronomical Algorithms and in Smart's Textbook on Spherical Astronomy, and we take the sign in which that point falls. John North's Horoscopes and History shows how astrologers have computed the ascendant since antiquity, and the definition has not changed.

The calculation works at every latitude where the ecliptic crosses the horizon, including the far north: our tests compute it for Tromsø, at almost 70° north. They also compare a birth in Rome on 20 May 1990 at 08:30 local time with an independent computation. At the poles the ascendant is undefined, and the calculator then shows no card rather than an invented one.

The ascendant uses the twelve tropical signs, so its card links to the pages of the Western signs.

The runic half-months divide the year into 24 periods of about fifteen days, each ruled by one rune of the Elder Futhark. The scheme on this site is the one set out by Nigel Pennick in Runes and Astrology, whose appendix gives the date and the time at which each half-month begins. It is a modern system, built by the author on the traditional order of the runes; we present it as Pennick's.

Because Pennick gives a starting time as well as a date, a birth on the day a half-month changes depends on the hour. His times are in local apparent solar time, the time shown by a sundial, so we convert your birth time to apparent solar time at your birthplace before comparing. This is why the calculator asks for the time and place for this sign.

Our tests check Pennick's boundaries one by one: Elhaz until 12 February at 09:59, Sigel from 10:00; the half-month of Eoh running across the turn of the year, from 28 December at 19:00 to 13 January; and the use of the solar date rather than the civil one near midnight. While checking them we found and corrected a mistake in the site's previous calculator, which had swapped two half-months, Odal and Dag.

In Burmese astrology a person's sign is the day of the week on which they were born. Each weekday has a planet, a direction and an animal: the Garuda for Sunday, the Tiger for Monday, the Lion for Tuesday, and so on. Wednesday is split in two: Wednesday morning belongs to the tusked elephant and Wednesday afternoon, from noon to midnight, to Rahu, the tuskless elephant, which gives eight signs in all.

The system is described by Shway Yoe in The Burman, His Life and Notions, and the reckoning of the day in Burmese astrology is discussed by J. C. Eade in The Calendrical Systems of Mainland South-East Asia. We follow the astrologers' rule that the day changes at midnight, and Shway Yoe's note that the English time system had come to prevail over the whole country; we therefore use the civil day and the local clock time of the birthplace.

This is why the calculator asks for the time and place: a birth on a Wednesday needs the hour, and a birth near midnight can fall on a different weekday in local time than in another time zone. Our tests check a Saturday, 1 January 2000, and the split of Wednesday 5 January 2000 at noon: 11:59 is the tusked elephant, 12:00 Rahu.

The lunar mansions, manāzil al-qamar, divide the Moon's monthly path into 28 stations, one for each night of its course. Each mansion has an Arabic name, often taken from the stars near it.

For astrological use the mansions were standardised as 28 equal parts of the zodiac, each of 12° 51′ 26″, counted from the beginning of Aries. This is the form given in the Picatrix, the great Arabic handbook of astral magic, and in al-Bīrūnī's Book of Instruction. We compute the Moon's apparent position at the instant of birth and take the mansion in which it falls: on 1 January 2000 at noon UT, for example, the Moon at about 223° was in the mansion al-Qalb, the Heart of the Scorpion, and on 4 July 2024 at 03:15 UT, at about 80°, in al-Dhirāʿ, the Forearm. Both are checked by our tests.

We present this result as a fact about the moment of your birth, "the Moon was in the mansion of…", and not as a sign with a personality: the classical sources use the mansions to choose good moments for actions, not to describe a person's character. Because the Moon moves through a mansion in about a day, the time of birth is always needed.

The Sefer Yetzirah, the oldest text of Jewish esoteric speculation, assigns each of the twelve signs of the zodiac to one of the twelve months of the Hebrew year, and the Jewish Encyclopedia describes how this pairing passed into later Jewish tradition. Modern kabbalistic astrology, for example Yehuda Berg's Kabbalistic Astrology, reads a person's sign from the Hebrew month of birth: Nisan is the Lamb (Aries), Iyar the Ox, and so on to Adar, the Fish.

We compute the Hebrew date of your birth with the arithmetic of the Hebrew calendar set out in Reingold and Dershowitz's Calendrical Calculations, and take the sign of its month. In leap years the calendar has two months of Adar; both belong to the Fish, as Berg gives it.

The Hebrew day begins at sunset, so a birth in the evening may belong to the next Hebrew date. We compute the sunset at your birthplace, at sea level and with the standard definition, the upper edge of the Sun's disc touching the horizon. Our sunsets agree with those of the US Naval Observatory to within a minute. On the last day of a month this decides the sign: on 8 April 2024 in Jerusalem, one minute before sunset is Adar II, the Fish, and one minute after is Nisan, the Lamb. Far north the Sun can set after midnight, so the Hebrew day begins at the last sunset before the birth, whatever the clock says; where there was no sunset at all in the day before the birth, as in the polar night, we show no sign rather than an invented one.

The Ethiopian calendar has twelve months of thirty days and a thirteenth month, Pagume, of five days, or six in the year before a leap year. Its year begins on 11 September, or on 12 September after a six-day Pagume, and it counts about seven years behind the Gregorian calendar. The sign on this site is the month of the Ethiopian calendar in which you were born, with Pagume as a thirteenth sign.

We compute the Ethiopian date with the arithmetic set out in Reingold and Dershowitz's Calendrical Calculations, and our calculation reproduces the sample dates printed there. The New Year moves as it should: 1 Maskaram 2016 fell on 12 September 2023, because the year before was a leap year, and 1 Maskaram 2017 on 11 September 2024.

We count the day from sunset, as in the Hebrew calendar. This is our convention, and we state it: the same rule for both calendars that begin the day in the evening keeps the calculation consistent. We compute the sunset at your birthplace at sea level, with the same definition we use for the Hebrew calendar, checked against the US Naval Observatory to within a minute. A birth after sunset on the last day of a month belongs to the next month, and where the Sun does not set we show no sign.

The Tibetan year animal changes at Losar, the Tibetan New Year, which follows the Phugpa calendar, the calendar of the main Tibetan tradition since the fifteenth century. It is a lunisolar calendar computed by arithmetic rather than by observation, derived from the Kālacakra Tantra, and Edward Henning's Kālacakra and the Tibetan Calendar describes it in full. Svante Janson's Tibetan Calendar Mathematics gives the same rules as exact formulas, which we follow.

The calendar is defined with fractions, so we compute it with exact rational arithmetic, never rounding, so that no day boundary can move by a rounding error. Our calculation reproduces the Losar dates we checked from Janson's table, from 1954 to 2027, the example given by Gantumur for 2027, and the Tibetan year of all 33 sample dates in Calendrical Calculations, from antiquity to the twenty-second century.

One more detail matters: in the Tibetan calendar the day begins at dawn, conventionally at 5:00 local mean solar time, as Henning explains, not at midnight. A birth in the early hours of Losar, before 5:00 by the Sun at the birthplace, still belongs to the previous year. This is why the calculator asks for the time and place: in Lhasa, 5:00 local mean solar time is about 06:55 Beijing time.

The Mongolian year animal changes at Tsagaan Sar, the Mongolian New Year. Mongolia follows its own version of the Tibetan calendar, the Tögs buyant system introduced in 1747, which uses the same kind of arithmetic as the Tibetan Phugpa calendar with different constants. As a result Tsagaan Sar and Losar usually coincide, but sometimes fall a day or a month apart: in 2025 Losar fell on 28 February and Tsagaan Sar on 1 March.

We follow the formulas given in Svante Janson's Tibetan Calendar Mathematics for the Mongolian calendar, and Gantumur's recent study of the Tibetan calendars, with exact rational arithmetic. Before admitting this system to the calculator we required it to reproduce the official Tsagaan Sar dates we had: it reproduces the dates we checked from Janson's table, from 1990 to 2030, including the control dates 1 March 2025 and 18 February 2026.

As in the Tibetan calendar, the day begins at dawn, at 5:00 local mean solar time, so a birth in the early hours of Tsagaan Sar still belongs to the old year. In Ulaanbaatar that moment falls at about 05:52 local time: our tests check that on Tsagaan Sar 2025 a birth at 05:50 is still in the Dragon year and one at 05:54 already in the Snake.

In Cambodia the animal of the year changes at Moha Songkran, the Khmer New Year in mid-April, at the moment when, by the traditional reckoning, the Sun enters Aries. Each year the Ministry of Cults and Religion announces the day and the time.

We compute that moment ourselves with the traditional method, as reconstructed by J. C. Eade in The Calendrical Systems of Mainland South-East Asia: the Sun's mean motion of the Khmer and Thai astronomical tradition, corrected by its equation of the centre, gives the instant at which the Sun's true longitude reaches the beginning of Aries, in Cambodian time. Our tests check Eade's own worked example and an instant he gives for 1898, and they reproduce the official date of Moha Songkran for every year from 2021 to 2026.

We state one convention clearly: the animal changes at the instant we compute, not at the time announced each year. The announced time can be some tens of minutes later than the computed one, and we have no rule to predict it. A birth within that short interval on the New Year's day could therefore be given the old animal by the official calendar and the new one by our calculator. Because the change happens at an instant, the calculator asks for the time and place of birth for this sign.

Ifá is the divination system of the Yoruba of south-western Nigeria, and one of the great oracles of the world: UNESCO inscribed it on the Representative List of the Intangible Cultural Heritage of Humanity in 2008. Its knowledge is held in 256 figures, the Odù, each with a large body of verses that a diviner, the babalawo, chants and interprets.

An Odù is not a birth sign. As William Bascom's classic study of Ifá divination describes, the babalawo casts sacred palm nuts or a divining chain, and the marks that fall build up, by chance, one of the figures; Bascom compares the method with the Chinese I Ching. The result reveals which Odù speaks to the question brought by the client at that moment. Bascom also records the Yoruba belief that each person receives a destiny at birth; Ifá is a way of learning about it through divination, not a rule that reads it from the date.

That is why we do not calculate it. No date of birth, however exact, can produce an Odù, and any calculator that claims to do so would be inventing a rule that the tradition does not have. The Yoruba pages of this atlas describe the Odù and their meanings; they do not assign one to you.

There is a genuine Sufi engagement with the heavens. Ibn al-ʿArabī, the great Andalusian mystic of the thirteenth century, built a cosmology in which the twenty-eight mansions of the Moon correspond to the twenty-eight letters, as William Chittick's study of his cosmology, The Self-Disclosure of God, explains. In this scheme the sphere of the fixed stars holds both the twelve constellations of the zodiac and the twenty-eight waystations of the Moon, and the interplay between the two sets the lower world in motion.

But Sufism has no system of birth signs. Ibn al-ʿArabī's cosmology is a map of being and of the soul's ascent, not a way of reading a person's character from the day of birth, and the classical Sufi tradition did not cast horoscopes of its own. The twelve signs presented on this site under the Sufi name are the ordinary signs of the zodiac seen through Sufi imagery, a modern presentation.

That is why we do not calculate a Sufi sign: there is no Sufi rule to apply. If you want the ordinary zodiac sign of your birth, the Western and Persian sections above give it exactly; the Sufi pages of this atlas describe the imagery and the ideas behind the name.

Ziwei doushu, "purple star astrology", is an authentic and old Chinese art of fate calculation. Its classical text survives in the Ming supplement to the Daoist Canon, the Xu Daozang of 1607, in three scrolls, and the Taoist Canon catalogue edited by Schipper and Verellen describes its place in that collection. Richard Smith's Fortune-tellers and Philosophers places it among the divination methods of late imperial China.

It is not a sign derived from a date, but a whole chart. From the date and hour of birth the practitioner sets up twelve palaces, such as those of life, wealth and marriage, and places in them many named stars, beginning with the Ziwei star itself, following the rules laid out in the text. The reading comes from the whole pattern, not from one position.

That is why we do not calculate it. Casting the chart correctly means applying many rules for which the text gives the method in compressed form, and we have no critical edition of those rules against which to check a calculation. Rather than offer a simplified "Ziwei sign", which the tradition does not have, we leave the chart to practitioners and describe the system on its pages. The calculator therefore shows no Ziwei card.

Ancient Egypt watched the sky with great care, but not through a zodiac of birth signs. Its own system was the decans, thirty-six stars or star groups whose risings at dawn marked ten-day intervals through the year and divided the night into hours, as Marshall Clagett's Ancient Egyptian Science describes. The twelve signs of the zodiac reached Egypt only late, from Babylon through the Greeks, and appear in Egyptian temples in the Ptolemaic and Roman periods, combined with the decans, as on the ceilings of the temple of Khnum at Esna and the temple of Hathor at Dendera, which Clagett publishes.

The "Egyptian zodiac" of twelve gods, in which a birth date gives you a god such as Isis or Anubis, has no ancient source. No Egyptian text assigns people a birth god from a set of twelve; the scheme is a modern popular invention.

That is why we do not calculate it: there is no ancient rule to apply, and the date ranges in circulation were made up recently. The Egyptian pages of this atlas present the twelve gods as the modern scheme they are, and describe the genuine Egyptian sky, the decans and the star clocks, in their history section.

The Druids of Iron Age Britain and Gaul left no writings. What we know of them comes from Greek and Roman authors and from much later Irish and Welsh texts, and none of these sources records a zodiac or a calendar of birth signs. Ronald Hutton's history of the Druids in Britain, Blood and Mistletoe, traces how the modern image of the Druid was built in the eighteenth and nineteenth centuries, much of it on the forgeries of Iolo Morganwg, who, as Hutton documents, invented triads and other texts and passed them off as ancient.

The "Druidic zodiac" of thirteen animal signs belongs to that modern tradition. Its thirteen periods follow exactly the month boundaries of Robert Graves's Celtic tree calendar, which Graves published in 1948 as his own reconstruction, and the versions in circulation differ from one book to another.

That is why we do not calculate it. There is no ancient rule, and no single modern one either, so any dates we chose would be our own invention. On this site the Celtic tree calendar, which has a precise modern author and precise dates, is calculated in its own section above; the Druidic pages describe the modern revival and its symbols.

The Inca read the sky in their own way. Besides the bright stars they recognised dark constellations in the dust lanes of the Milky Way, a llama, a fox, a toad, a serpent, and they followed the Pleiades and the Sun through the year to time the agricultural calendar. Gary Urton's study of Andean astronomy, At the Crossroads of the Earth and the Sky, describes this sky as it is still read in the Andes: the dark constellations are called yana phuyu, dark clouds, and the Pleiades are watched to time the cycle of maize.

The Inca kept a calendar of months and festivals, but nothing in the Andean sky lore that Urton records works as a zodiac of birth signs. The idea of being born under one of twelve Inca signs, read like a horoscope, is a modern arrangement laid over the Andean sky lore.

That is why we do not calculate it: there is no Inca rule that turns a date of birth into a sign, and the date ranges in circulation are recent inventions. The Inca pages of this atlas present the twelve signs as the modern scheme they are and describe the genuine Andean astronomy behind them.

Aboriginal and Torres Strait Islander Australia is not one culture but hundreds of nations, each with its own language, Law and knowledge of the sky. That knowledge is among the oldest astronomy on Earth. Duane Hamacher's The First Astronomers, written with Elders and knowledge holders, describes how the rising and setting of stars mark the seasons, and how dark shapes in the Milky Way, such as the Emu in the Sky, signal when to gather emu eggs: the eggs are collected from April to June, and the celestial emu changes meaning with the seasons, from a female running to a male sitting on the nest.

None of these traditions is a zodiac of birth signs. There is no pan-Aboriginal calendar that assigns a person a sign from the date of birth, and the twelve-sign "Aboriginal zodiac" sometimes found in popular books has no source in the traditions themselves.

That is why we do not calculate it. The sky knowledge of each nation belongs to that nation, and it is not ours to turn into a horoscope. The Aboriginal pages of this atlas present the twelve signs as a modern overlay and describe the genuine astronomy behind them with the respect it deserves.

Traditional Hawaii kept a lunar calendar. Each month had thirty named nights following the phases of the Moon, each with its own character for fishing and planting, and the year of twelve months was kept in step with the seasons by inserting an extra month about every third year, by a rule that is not known. David Malo's Hawaiian Antiquities, written in the nineteenth century by a Hawaiian scholar and translated by Nathaniel Emerson, records the months and the nights of the moon, from Hilo, the first night, to Mahealani, the full moon of the sixteenth night, and the great festival of the Makahiki.

The calendar is real, but a zodiac of thirteen personality signs, one per month, read like a horoscope, is not part of what Malo or the other traditional sources record. It borrows the thirteen months and gives them horoscope meanings, a modern reinterpretation.

That is why we do not calculate it. The traditional months were lunar and moved against our calendar from year to year, so fixed date ranges cannot represent them, and the personality signs have no traditional source. The Hawaiian pages of this atlas present the thirteen signs as the modern scheme they are, and describe the moon calendar behind them.