Heliacal Risings

A star near the ecliptic disappears for part of each year, lost in the Sun’s glare. Its heliacal rising is the first morning it becomes visible again in the dawn twilight before sunrise; its heliacal setting is the last evening it can be caught in the dusk. Both events are annual, both repeat reliably, both are observable without instruments, and both depend on latitude. Before mechanical timekeeping and before calendars that could be trusted to stay in step with the seasons, this was the best annual clock available to anyone.

The earliest European use of it

The oldest European text organised around stellar phases is Hesiod’s poem on farming and the year, from about 700 BCE. Its instructions are keyed to the sky rather than to a calendar: begin the harvest when the Pleiades rise, and the ploughing when they are setting; the Pleiades are gone for about forty days between the two; prune the vines when Arcturus rises at dusk sixty days after the winter solstice; Sirius is at its most oppressive in high summer, when the goats are fattest and the wine is sweetest ¹.

It is a working almanac using the only reliable annual index a farmer had, in a world where every city kept a different calendar and magistrates inserted intercalary months when it suited them.

Parapegmata

The genre this becomes is the parapegma: an inscribed stone or written text correlating stellar phases with weather predictions and, in some cases, with civil calendar dates. A physical parapegma had a row of holes and a movable peg, advanced daily, so the reader could find the current position by inspection ².

Parapegmata are attested from the fifth century BCE and survive in inscribed examples and embedded in the Roman agricultural writers, one of whom includes a complete one in a book on farming ². They are the closest ancient analogue to what a location-aware seasonal calendar does now: take a date and a place, report what the sky is doing, and say what it implies for practical life. That resemblance locates this kind of computation in a tradition older than astrology and independent of it.

Why the dates drift

The calendar date of a given star’s heliacal rising moves, and it moves for the same reason the stellar and seasonal zodiacs separate: precession carries the equinox westward along the ecliptic at about 50.3 arcseconds a year ³. That works out at roughly one day per seventy-one years, which is about a month since Hesiod. See Precession of the Equinoxes.

The consequence is concrete. A reader who follows Hesiod’s instruction literally today, and begins the harvest at the heliacal rising of the Pleiades, is harvesting about a month later in the seasonal year than he intended. The instruction was correct when written, is precisely recorded, and has been overtaken by the sky. That is a more instructive fact about the limits of literal traditionalism than any argument could be.

Correction

Commonly saidA heliacal rising can be computed from the geometry: the star is up before the Sun, so it is visible.

Geometry gives a necessary condition only. Whether a star of a given magnitude can be seen at a given altitude in a given twilight depends on atmospheric extinction along a long slant path, on the brightness of the twilight sky, on the observer’s visual acuity, and on the horizon and the weather at the site. A star can be geometrically above the horizon and entirely invisible.

The standard treatment layers a visibility model, with an extinction coefficient and an arcus visionis criterion, on top of the positional computation ⁴. The difference between a geometric estimate and a modelled one runs to days.

The options are therefore two: implement the visibility model, or state the simplification in the interface where the number is shown. This site does not present a geometric date as a heliacal date without saying so.

The same physics, in the sky view

The three-dimensional sky view on the chart draws stars as they would have looked from the birthplace, which means it needs the same quantities this entry has been describing, and each has a source.

Brightness follows the magnitude scale itself, which is a ratio: five magnitudes are a factor of one hundred in received light, so one magnitude is the fifth root of a hundred, about 2.512. Norman Pogson proposed exactly that figure, choosing it so the fifth power would come out at a round hundred ⁵. A star two magnitudes brighter than another is about six and a third times as bright.

Dimming near the horizon follows airmass, the length of the slant path through the atmosphere relative to straight up. The naive formula for this is the secant of the zenith angle, which is fine overhead and diverges to infinity at the horizon, where the real answer is finite and near forty. The view uses the revised formula of Kasten and Young, which is well behaved all the way down ⁶. The extinction coefficient applied to that path is a typical value, because the weather at a birth is not recoverable, and Schaefer’s per-site table is the reference for what a plausible range looks like ⁴.

Extinction is wavelength dependent, and that dependence is the same one that makes the daytime sky blue: scattering rises steeply as wavelength falls, so blue light is removed from a slant path more strongly than red ⁷. A star low in the sky is therefore fainter and redder, which is why a setting star looks the way it does and why the view reddens as well as dims them.

What this makes computable

For a given latitude and year, the heliacal rising and setting dates of the bright ecliptic and near-ecliptic stars can be computed and compared with the dates a named ancient text assigns them, showing the precessional drift directly. The same machinery gives the circumpolar limits for a latitude, which stars never rise and which never set, and the twilight behaviour that decides whether a heliacal observation is possible at all: north of about 48 and a half degrees, astronomical twilight never ends in late June, so there is no fully dark sky at midsummer in Paris or Vienna.

That last fact explains a great deal. Midsummer means something different at the latitude of Riga than at the latitude of Palermo, and the difference is measurable. See Solstices and Equinoxes and Cross-Quarter Days.

Gap in the evidence

Whether stellar phases contributed to the salience of the early November and early May positions in northern Europe is unknown. The Pleiades set heliacally in early November and rise heliacally in early May at Mediterranean latitudes, and they are the most widely used calendar asterism on Earth. No northern European source connects them to the cross-quarter stations, the dates of the phases shift with latitude and with precession, and the coincidence is exactly the kind that the older archaeoastronomy would have built a system on ⁸. It is recorded here as a hypothesis and is not adopted.

References

Hesiod, Glenn W. Most (ed.), trans. Glenn W. Most. Hesiod, Volume I: Theogony, Works and Days, Testimonia. Loeb Classical Library 57. Cambridge, Massachusetts: Harvard University Press, 2006 (composed c. 700 BCE). ISBN 9780674996229 recordprimary source
Works and Days is the earliest European star-calendar: the heliacal rising and setting of the Pleiades for harvest and ploughing, Arcturus sixty days after the winter solstice, Sirius in high summer. A revised second edition of 2006 appeared in 2018 under a different ISBN and is a distinct book.
Daryn Lehoux. Astronomy, Weather, and Calendars in the Ancient World: Parapegmata and Related Texts in Classical and Near-Eastern Societies. Cambridge: Cambridge University Press, 2007. ISBN 9780521851817 recordpeer reviewed
The definitive study of the parapegma, the star-and-weather almanac genre, with a catalogue and translations. Reviews render the subtitle variously as "Near-Eastern" and "Ancient Near Eastern"; the form here follows the publisher's own record.
Bradley E. Schaefer. Astronomy and the Limits of Vision. Vistas in Astronomy, 1993, 311-361. doi:10.1016/0083-6656(93)90113-X recordpeer reviewed
Volume 36. The standard visibility and atmospheric-extinction model for heliacal phenomena; a geometric altitude test alone will not predict first or last visibility of a star.
Jean Meeus. Astronomical Algorithms. Richmond, Virginia: Willmann-Bell, 1998. ISBN 0943396611 recordscholarly
Second edition, 1998. The standard reference for the algorithms behind solstice, equinox and solar-longitude computation, including the equinox and solstice chapter this site's seasonal calculations follow.
Clive L. N. Ruggles. Astronomy in Prehistoric Britain and Ireland. New Haven: Yale University Press, 1999. ISBN 0300078145 recordpeer reviewed
The model of post-reform archaeoastronomical method: a defined site sample, all orientations recorded, and statistical testing against a null of random orientation. Yale University Press carries both New Haven and London imprints and reviews cite the book under each; New Haven is used here.
Norman R. Pogson. Magnitudes of Thirty-six of the Minor Planets for the First Day of each Month of the Year 1857. Monthly Notices of the Royal Astronomical Society 17(1), 1856, 12-15. doi:10.1093/mnras/17.1.12 record Accessed 2026-07-31.primary source
The original source of what is now called Pogson's ratio, 2.512, the fifth root of 100. Bibliographic details (MNRAS vol. 17, issue 1, pp. 12-15, 14 November 1856) verified against the Oxford Academic publisher record for the article and against ADS bibcode 1856MNRAS..17...12P. The DOI resolves to the same OUP page. Read in full from the ADS scan (not a summary): the paper is mainly a table of predicted opposition magnitudes for 36 minor planets for 1857, used to avoid mistaking a nearby star for the object sought. The magnitude scale is introduced only in passing, to explain how the table's monthly columns were computed: "calculated on the assumed ratio of the light of 2.512, i.e. that a star of any magnitude, as for instance the eighth, contains 2.512 times the light of the next less, or ninth magnitude" (p. 13). Pogson goes on to explain the choice (p. 14): earlier observers had derived ratios scattered between about 2.4 and 2.83 (Dawes' 4, Johnson's 2.43, his own independent mean of 2.4, Steinheil's 2.83, Stampfer's 2.519), and among these "it signifies little which... is adopted"; he "selected 2.512 for convenience of calculation," because it is the value whose reciprocal of one-half its logarithm is exactly 5: in modern terms, the ratio whose fifth power is exactly 100. He does not use the word "magnitude scale" as a named system; that framing, and the name "Pogson's ratio," are later usage. This passage was verified against the page image itself (300 dpi scan), not the OCR text layer, because the OCR renders the key fraction ambiguously.
Fritz Kasten, Andrew T. Young. Revised optical air mass tables and approximation formula. Applied Optics 28(22), 1989, 4735-4738. doi:10.1364/AO.28.004735 recordpeer reviewed
The standard closed-form approximation for optical air mass as a function of (apparent) zenith angle, valid all the way to the horizon, where the naive sec(z) diverges. Bibliographic details verified directly against the publisher record at opg.optica.org (Optica Publishing Group, formerly the Optical Society), cross-checked against pveducation.org's citation of the same paper, and the DOI resolves. This is the formula generally meant by "the Kasten-Young airmass formula" in solar-energy and atmospheric-optics software (e.g. pvlib). See also young-1994-air-mass-refraction, a later refinement by the same second author for use with true (unrefracted) rather than apparent zenith angle.
Lord Rayleigh (John William Strutt). On the Light from the Sky, its Polarization and Colour. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science 41(271), 1871, 107-120. doi:10.1080/14786447108640452 recordprimary source
The original derivation of the inverse-fourth-power wavelength dependence of scattering by particles small compared to the wavelength of light, the basis of "Rayleigh scattering" and the reason the daytime sky is blue. Read in full via the Wikisource transcription of Rayleigh's collected Scientific Papers, which reproduces the original text: "the ratio of the amplitudes of the vibrations of the scattered and incident light varies inversely as the square of the wave-length, and the intensity of the lights themselves as the inverse fourth power." Volume, issue and page range (Phil. Mag. Ser. 4, vol. 41, pp. 107-120) cross-checked against the table of contents of "Scientific Papers, Vol. 1: 1869-1881" and independent citation records; the DOI resolves (redirecting to the Taylor & Francis article page, which blocks automated fetches directly). Note this is Part I of a two-part paper; Part II continues at pp. 274-279 of the same volume and is not separately verified here.