Little Almanac

Three Reference Frames

Three different systems assign a name to a position on the ecliptic. They are built on different anchors, for different purposes, by people separated by two thousand years, and they give different answers because they are answering different questions. Almost every public argument about “what sign someone really is” comes from treating them as three rival answers to one.

The tropical frame: a seasonal grid

Zero degrees of Aries is the March equinox point, where the ecliptic crosses the celestial equator and the Sun passes from south to north. From there, twelve equal arcs of thirty degrees. Zero Cancer is the June solstice, zero Libra the September equinox, zero Capricorn the December solstice. The cardinal signs are, by construction, the season-openers.

Its relation to the fixed stars is none, by definition. It is anchored to the geometry of the Earth and the Sun.

This matters for a claim that circulates constantly: that the tropical zodiac has drifted. Measured against the seasons, which are what it is anchored to, it is fixed permanently. Measured against the stars, the whole frame moves at about 50 arcseconds a year, which is precession itself.

The convention was adopted explicitly, with a stated reason, by the second century CE at the latest. Ptolemy grounds the choice in the fact that the equinoctial and solstitial points are naturally determined whereas the constellation figures have no definite boundaries 1 bk. I ch. 11, and Geminos, writing earlier, already distinguishes the twelve equal signs from the constellations that share their names 2. Both authors knew about precession. Hipparchus had described it in the second century BCE, comparing his own measured stellar longitudes with observations made roughly a century and a half earlier and concluding that the rate was not less than one degree per century; the report survives through the Almagest 3 bk. VII chs. 2-3.

The sidereal frame: a star-referenced grid

Twelve equal thirty-degree signs again, but with the zero point fixed by reference to the stars rather than to the equinox. The offset between the two zero points is called the ayanamsha, and it has no uniquely correct value: it must be stipulated. Values in current use differ by roughly two degrees, depending on which star is taken as the anchor and what longitude it is assigned.

The two frames coincided at some point around the third century CE, and the timing matters for reading Ptolemy. Within his lifetime the tropical and sidereal zodiacs agreed to within a degree or two, depending on the convention. He was not choosing between two visibly different systems, and the divergence became conspicuous only much later 4.

This frame is the one used by Babylonian astronomy, by the Indian astrological tradition, and by a small Western sidereal school founded in the twentieth century. It is also the only frame in which “a different sign” is a coherent statement, because it is the only one making a claim about where the stars are.

Gap in the evidence

Whether the sidereal frame has been used continuously by Indian astrology across roughly two millennia is a claim about unbroken transmission, and nothing in this bibliography supports or refutes it: the sources listed here are European-language works on the Western tradition, and no Indological scholarship has been consulted.

What can be said without an Indological source is narrower. Sidereal reckoning is the frame in which the Indian tradition works, and the offsets this site implements are named for the people who fixed them. How that frame was transmitted, when, and with what continuity is a question this site is not equipped to answer.

The IAU constellations: a cataloguing convention of 1930

Eighty-eight regions partitioning the entire celestial sphere with no gaps and no overlaps, bounded by arcs of constant right ascension and constant declination referred to the equinox of B1875.0. They were drawn by the Belgian astronomer Eugène Delporte for the International Astronomical Union: proposed at the General Assembly at Cambridge in 1925, approved at Leiden in 1928, published in 1930 5.

Their purpose was to give every star and every point on the sky an unambiguous address, so that a variable star or a nova could be named and catalogued without dispute. They are administrative regions, closer to postcodes than to pictures.

That last point has a consequence which surprises people who assume the constellations are more official than they are. The IAU fixed the boundaries and never fixed the figures. The lines joining Orion’s stars into a hunter, or the seven stars of the Plough into a shape, are a drawing convention with no governing authority behind them, and different atlases join different stars. The dataset this site uses to draw them takes its lines from the IAU’s own published charts and then records that its compiler modified some of them 6, which is an accurate statement of the position: there is no canonical answer for it to have departed from. The star positions underneath are a different matter and are firmly sourced, coming from the Hipparcos reduction by way of the XHIP compilation 7.

So of the three things a constellation might mean, only two are officially settled: where its boundary runs, and where its stars are. What it looks like is convention.

A convention with a date on it

The convention has an origin, and the sky view can show it. Johann Bayer’s Uranometria, printed at Augsburg in 1603 and engraved by Alexander Mair, was the first atlas to map the whole sky on a uniform plan, and it introduced the scheme of lettering a constellation’s stars in Greek 8. Those are the Bayer designations, and they are still in use: the alpha in Alpha Centauri is his. A modern catalogue and a four-hundred-year-old engraving therefore label the same stars with the same letters, which is what makes it possible to lay one over the other.

Three of the plates are aligned to the real sky in the chart’s sky view, from the ETH-Bibliothek’s public domain scan 9. The alignment is a fit, so it has an error, and the error differs from plate to plate.

Established

Fitting each plate to the modern positions of its own lettered stars gives a median disagreement of about 0.22 degrees for Orion and 0.19 for Taurus, which is close to the limit of what can be read off an engraving. For Ursa Major it is 1.41 degrees, six times worse, and the reason is visible on the plate: the engraver drew the bear’s tail as a composed curve, and the stars went where the picture needed them.

That single number separates an atlas as a measuring instrument from an atlas as a picture with stars in it. Uranometria is both, and it is more one than the other depending on which page is open.

There is also something the fit cannot do, and the attempt is the obvious next thought. Bayer’s positions came from Tycho Brahe’s catalogue, for an epoch around 1600, and the modern positions are for 2000, four centuries of precession apart. It is tempting to think the plate would fit its own epoch better and that the residuals could therefore date it. They cannot. Precession is a rotation of the whole sphere, and a rotation preserves every angle between every pair of stars, so a fit that is free to rotate absorbs four hundred years of it without leaving a trace. Both epochs give the same answer to three decimal places. Better measurement cannot date a plate from its stars alone, because the information is not in the pattern. It is in the coordinate figures printed around the border, which is a different reading task.

Four properties follow, and all four matter here.

  • They are wildly unequal in size, and along the ecliptic the inequality is extreme. Virgo spans nearly forty-four degrees of ecliptic longitude; Scorpius spans under seven.
  • They are not the star patterns. The boundary lines are rectilinear in an 1875 coordinate grid; the figures are a separate, older and never-standardised tradition.
  • They were never designed as, intended as, or usable as a reference frame for assigning meaning to positions.
  • The ecliptic passes through thirteen of them, because the region assigned to Ophiuchus extends across the ecliptic between Scorpius and Sagittarius.

A figure such as “Virgo spans nearly forty-four degrees” is not a number Delporte published. He drew the boundaries as arcs of right ascension and declination referred to the equinox of B1875.0 5; the ecliptic-longitude widths quoted for each constellation across this site are this site’s own computation, found by crossing those boundaries against the ecliptic, and are cited to Delporte only for the boundaries themselves, never for the derived decimal figure.

Correction

Commonly saidBecause of precession there is a thirteenth sign, Ophiuchus, and most people's sign is really the previous one.

What is correct in the claim, and should be conceded first: precession is real, the accumulated offset between the tropical zero point and the stars is now roughly twenty-four degrees, and for most people the constellation the Sun stood in front of at their birth is not the constellation their tropical sign is named after. Astrologers’ signs and astronomers’ constellations have not lined up for a long time.

What is confused in it can be shown definitionally, without settling whether astrology works.

A tropical sign is a thirty-degree arc measured from the March equinox. It marks seasonal position and says nothing about which stars lie behind the Sun, a separation in force since Ptolemy at the latest 1 bk. I ch. 11. Testing a tropical sign against constellation membership applies the definition of one system to the output of another.

The instrument is also the wrong one. The IAU boundaries are a 1930 cataloguing convention drawn for a purpose unrelated to the question 5, and adopting them imports a set of rectilinear boundaries into systems that have always used equal divisions.

And there is no thirteenth sign in any astrological system. Sidereal astrology, the school for which the star-relative position is the point, still uses twelve equal signs and does not include Ophiuchus. Reaching a thirteenth requires abandoning both astrological frames.

Finally, it is not news. Precession has been known for about 2,100 years, the tropical convention was adopted by authors who knew it, and the twentieth century saw a full internal debate about which frame to use 4.

The best-known outbreak of the story followed remarks in mid-January 2011 by Parke Kunkle, an astronomy instructor and a board member of a planetarium society. Two points of fairness are owed here: the observation he made about precession and constellation boundaries was ordinary and correct, and he did not claim that anyone’s sign had changed. The framing that signs had changed was added by the coverage.

Gap in the evidence

The date of the originating newspaper piece is not settled here. Sources differ between two days in mid-January 2011 and the original is paywalled, so this entry says “mid-January 2011” rather than choosing.

Separately, why Greek astronomy adopted the tropical convention when its Babylonian source was sidereal has not been established. The available answers, that it followed from Greek interest in the solar year, that Ptolemy’s stated argument about the indefiniteness of constellation figures is the real reason, and that the shift was gradual and semi-conscious because the two frames nearly coincided in the relevant centuries, are all defensible, and no one has shown which was decisive.

What the three frames are each good for

The tropical frame answers: where is the Sun in the cycle of the seasons? That question is definitionally exact and no observation can contradict it.

The sidereal frame answers: where is a body against the fixed stars, on a chosen convention? That question is answerable once an ayanamsha is named, and any careful presentation names it.

The IAU frame answers: which catalogued region of sky does a body occupy? That question has one settled, falsifiable, observational answer, and it is the only one of the three that can be checked with a telescope.

Whether any of the three supports interpretive claims built on it is a separate question, which this page leaves alone. See The Tetrabiblos for the text in which the tropical convention is argued, and Virgo for what happens to a sign’s identity across two of these frames and four layers of history.

References

  1. Claudius Ptolemy, trans. F. E. Robbins. Tetrabiblos. Loeb Classical Library 435. Cambridge, MA: Harvard University Press, 1940 (composed c. 150 CE). bk. I ch. 11. find a copy (catalogue search) recordprimary source

    The standard English reference and the chapter-numbering scheme this site adopts and names. Non-Robbins reprints number the chapters differently, so every locus here is given as Robbins numbering.

  2. Claudius Ptolemy, trans. G. J. Toomer. Ptolemy's Almagest. London: Duckworth, 1984 (composed c. 150 CE). bk. VII chs. 2-3. find a copy (catalogue search) recordprimary source

    Reissued by Princeton University Press, 1998. Book VII chapters 2-3 for precession.

  3. Eugène Delporte. Délimitation scientifique des constellations: Cambridge University Press for the International Astronomical Union, 1930. find a copy (catalogue search) recordprimary source

    The 88 constellation boundaries, drawn as arcs of constant right ascension and declination referred to the equinox of B1875.0. Chronology: proposed at the IAU General Assembly at Cambridge in 1925, approved at Leiden in 1928, published in 1930.

  4. Geminos, trans. James Evans, J. Lennart Berggren. Geminos's Introduction to the Phenomena: A Translation and Study of a Hellenistic Survey of Astronomy: Princeton University Press, 2006 (composed 1st c. BCE). find a copy (catalogue search) recordscholarly

    Geminos distinguishes the twelve 30-degree signs from the constellations of the same names and corrects the popular confusion directly. This site cites the book for that distinction but does not quote the passage: the chapter and section reference in circulation is corroborated only from secondary sources quoting the translation, and has not been checked against the printed text.

  5. Nicholas Campion. A History of Western Astrology, Volume II: The Medieval and Modern Worlds: Continuum, 2009. ISBN 9781441181299 recordscholarly

    The companion volume, and the one that carries everything after the fifth century: its own description runs from the collapse of classical astrology to popular astrology in the twentieth century and its standing in the twenty-first. Volume I stops in the ancient and classical world, so a claim about the Renaissance, the Victorian press, the newspaper column or twentieth-century psychological astrology belongs here and not there. Listed by Bloomsbury Academic since Bloomsbury acquired Continuum in 2011; the 2009 printing is Continuum, and the ISBN is unchanged.

  6. Olaf Frohn. d3-celestial: star, constellation-line and star-name data files, 2015. https://github.com/ofrohn/d3-celestial record Accessed 2026-07-30.primary source

    Source of both datasets behind the sky view: `data/stars.6.json` and `data/stars.8.json` (J2000 equatorial positions, Hipparcos number, V magnitude and B-V colour index, converted from VizieR V/137D, the XHIP compilation, see anderson-francis-2012-xhip), `data/starnames.json` (proper name, Bayer and Flamsteed designations, compiled from VizieR IV/27A and IV/22), and `data/constellations.lines.json` (the 88-figure stick-figure line data, per the repository's own readme.md "Sources" section drawn from the IAU's public constellation charts with line modifications by Frohn himself). The whole repository, including the `data/` directory, is released under the 3-clause BSD licence in its root `LICENSE` file (copyright 2015 Olaf Frohn); verified by fetching that file directly. A BSD licence permits redistribution of a modified, compacted form with the copyright notice retained, which this project does in scripts/build-sky.ts and public/sky/manifest.json. Data fetched at the `master` branch commit 6d5e96655e0f1fe70b1eb7cf9be0eed5290e1458 (the most recent commit touching `data/stars.6.json`, dated 2020-02-29). Rejected on licensing grounds before this source was chosen: the HYG Database (astronexus/HYG-Database), whose LICENSE file is Creative Commons Attribution-ShareAlike 4.0, a copyleft term this non-share-alike site cannot take on for a data file it compacts and redistributes; and the Stellarium `western` sky culture (Stellarium/stellarium-skycultures), whose per-culture `index.json` carries no `license` field of its own, the repository root license is the copyleft AGPL-3.0, and the README only recommends, without requiring, CC BY-SA for individual cultures' content, which leaves the actual terms for that specific file unstated.

  7. E. Anderson, Ch. Francis. XHIP: An extended Hipparcos compilation. Astronomy Letters 38, 2012, 331-346. doi:10.1134/S1063773712050015 recordpeer reviewed

    The scientific origin of the star positions, magnitudes and colour indices in the sky-view catalogue: every entry in the New Reduction of the Hipparcos Catalogue, cross-referenced against a broad set of other catalogues (Bayer/Flamsteed/variable-star identifiers, radial velocities, spectral types). Published as VizieR catalogue V/137D, whose data are free for scientific use with citation (https://cds.unistra.fr/vizier-org/licences_vizier.html). This site does not fetch V/137D directly; it uses ofrohn-d3-celestial (see that source), which converts XHIP into a compact GeoJSON star list under its own BSD-3-Clause licence. Citation verified against ADS bibcode 2012AstL...38..331A and the Semantic Scholar record for the DOI above, both matching the arXiv preprint 1108.4971.

  8. Johann Bayer. Uranometria: omnium asterismorum continens schemata, nova methodo delineata aereis laminis expressa. Augsburg: Christophorus Mangus, 1603. https://doi.org/10.3931/e-rara-309 record Accessed 2026-07-31.primary source

    The first atlas to cover the whole sky systematically, and the origin of the Bayer designations, the Greek-letter star names this site's own catalogue still carries (see public/sky/stars.json's `bayer` array). 51 engraved plates, one per constellation plus two planispheres, engraved by Alexander Mair (1559-1617); the imprint names Christophorus Mangus, though the copy digitised here records the actual print shop as Sara Mang's Augsburg press. VD17 39:125032X (the German union catalogue of seventeenth-century imprints). Three plates (Orion, Ursa Major, Taurus) were aligned to modern star positions for the sky overlay in scripts/build-atlas.ts; see eth-bibliothek-uranometria-scan.yaml for the scan this project used, which is a separate licensing question from the 1603 work itself being long out of copyright.

  9. ETH-Bibliothek Zürich. Ioannis Bayeri Uranometria omnium asterismorum continens schemata (digitised copy, ETH-Bibliothek Zürich, Rar 8931), 2010. https://doi.org/10.3931/e-rara-309 record Accessed 2026-07-31.primary source

    The digitisation actually used for the three aligned plates (Orion, Ursa Major, Taurus) in public/atlas/, as a licensing question separate from the 1603 work itself: see bayer-1603-uranometria.yaml for the atlas. Fetched at 300 dpi via e-rara's own PDF export (https://www.e-rara.ch/zut/download/pdf/77412) and IIIF image service (https://www.e-rara.ch/i3f/v20/77412/manifest, canvas dimensions confirmed identical to the PDF's embedded raster, so nothing higher-resolution was available to fetch). LICENCE, verified directly: the record page states the item is marked Public Domain, and e-rara's own terms-of-use page (https://www.e-rara.ch/wiki/termsOfUse) states its digitisations may be used freely for scientific, private, non-commercial and commercial purposes, requesting only source attribution (institution, holding shelfmark and the persistent DOI) in return, which this project's manifest.json and the note above both carry. CANDIDATES CHECKED AND REJECTED before this one: - Wikimedia Commons (e.g. File:ORION_Uranometria_orion.jpg) states the image is "courtesy of the United States Naval Observatory Library, who gives explicit permission to use it so long as attribution is attached", but that permission is an unlinked claim by the uploader with no cited permission letter, URL or other traceable document, and a search of the USNO Library's own site found no digitisation project or public statement corroborating it. That is exactly the "Commons file's own licence tag is a claim by an uploader" case this project's own standard requires tracing to the holding institution rather than accepting; since it could not be traced, it was not used. - The Library of Congress holds a copy (Lessing J. Rosenwald Collection) but the specific digitised item page returned HTTP 403 to an automated fetch and no IIIF/direct-image access was confirmed; it was left unpursued once e-rara's terms were already confirmed favourable and its IIIF access already working. - The Linda Hall Library's "Out of This World" exhibition displays Uranometria pages but as a curated exhibition, not a full-plate archival image service with a stated reuse licence for individual pages; not checked further for the same reason as the Library of Congress. e-rara/ETH-Bibliothek was chosen because it is the only candidate whose reuse terms for this specific item were confirmed in writing on an institutional page, with a working, unauthenticated, full-resolution image service.